As the virtualization boom continues, many practitioners are looking to expand beyond the server and into other aspects of their architecture.  Their intent is simple: if server virtualization provides such flexible and simple deployment, then virtualizing other aspects of the network should provide even more flexibility and simplicity; you could run an entire application segment, including switches, routers and firewalls all with the confines of a single physical hardware device. Moving that application segment to a new datacenter or backup facility would be as simple as copying files. Consequently, many manufacturers of network and application delivery equipment have rushed virtual versions of their products to market; some completely replacing their physical counterparts and others providing scaled down virtual versions in order to have a product, but not one that cannibalizes their traditional product sales.

Unfortunately, many of these offerings were developed simply in the face of virtualization hype and little thought seems to have been put into exactly how these devices would or should work, where it might make sense to use a virtual appliance in lieu of a physical one or when a physical appliance still makes the most sense.  There are certainly many cases where virtual instances of physical appliances make perfect sense: development and testing labs, for training purposes and even for proof-of-concept deployments for new projects.  All of these situations can be vastly improved by the ease of deployment, lower cost and flexibility offered by virtual solutions, but there is still a lot of debate about their place in realm of production deployment.

I have to say – and I have said it before – I am not a great fan of the ‘virtual appliance’ model for delivering enterprise management software. Specifically, I have ongoing concerns about how these software appliances break compliance, security, and other important management and policy requirements. – Andi Mann,Virtual Appliances – More Risk than Reward?, October 29, 2009.

Of course this doesn’t even take into account that many physical appliances started out on general computing hardware and evolved over years to include special purpose silicon and other unique hardware in order to achieve the speed an performance required by today’s applications; many of which simply aren’t available to virtual incarnations.

It seems that there is definitely a place in the enterprise for virtual versions of network and application delivery appliances, but there are some basic requirements that must be met.  The virtual version must be completely compatible with its physical counterparts and have essentially the same capabilities—this makes it easy to use virtual instances in dev/test and move configuration files in bulk to their physical counterparts in production; the corollary being that virtual versions are still likely to be compliments to their physical brethren, not wholesale replacements. Finally, given the constantly changing needs of the environments mostly likely to benefit from virtual appliances, there needs to be a great deal of flexibility and leeway in licensing that meets this need.

 F5 recently announced availability of a free trial version of its new virtual appliance, the BIG-IP^® Local Traffic Manager^™ (LTM^™) Virtual Edition (VE). This virtual version of F5’s flagship application delivery control helps address the immediate needs of customers while not eliminating their existing appliance investment or relegating them to only basic functionality; giving customers a choice of the how these devices best fit within their environment.

Developers can now easily work directly with the services provided by the LTM while developing applications, integrating iRules and iControl components to take advantage of centralized network services.  These components can be easily moved with the applications directly into production because the LTM VE is an integrated component of the overall application delivery network, not simply a one-off solution.  This includes free access to the support portal and resources on DevCentral to make it easier to break the traditional barriers between development and the network.

While application architects and developers don’t necessarily care about the final implementation of ADC services, they require a flexible infrastructure that can easily be integrated into the application development and deployment process, and can adapt quickly to business operations and opportunities. Soft ADCs give customers the flexibility to deploy ADCs in whole new ways to test, develop, and integrate more tightly into an overall Enterprise Cloud Architecture. - Mark Fabbi, VP and Distinguished Analyst at Gartner

In addition, F5 is now offering free BIG-IP LTM web-based training as a flexible learning tool to help users get the most from F5 solutions.  This makes the LTM VE a perfect vehicle for development, testing and proof-of-concept use, particularly with organizations who may be unfamiliar with the benefits and breadth of services offered in today’s application delivery solutions.  It also minimizes the cost of education and training for organization standardizing on F5 technology.

This is simply the first step towards an overall Enterprise Cloud Architecture strategy that began back in 2007 with the introduction of F5’s unique Clustered Multiprocessing (CMP) technology. This strategy and vision was recently bolstered by the new BIG-IP Edge Gateway^™ and BIG-IP Access Policy Manager^™ products, which can serve as application access control points for internal and external enterprise cloud services. These solutions simplify and unify application access control while ensuring high levels of quality of service.  Tight integration between physical appliances that can provide consistent, reliable control, virtual appliances which can easily be moved between datacenters and cloud peering-points and advanced services which ensure the availability, security and performance of applications is a critical component of future cloud development.

Press Release
 
Support for BIG-IP LTM VE

 Server virtualization is booming.  Enterprise organizations and mom-n-pop operations alike are experiencing a whole new level of flexibility within their IT operations.  Server virtualization not only provides cost savings through consolidation, but also provides reduced deployment times and greater control over the use of finite resources.  The sophisticated tools provided by today’s leading hypervisor vendors create the ability to dynamically balance the resource needs of applications across the available resources.  Instead of having to invest in resources to meet the needs of every application’s peak requirements, they can dynamically move unused resource to augment applications experiencing peak traffic. For instance, accounting systems often need significantly more processing resources at the end of the month, quarter or year than they do to handle day-to-day operations.  Instead of having to deploy static resources to meet the needs of this peak usage, as traditionally done, server virtualization allows you to move workloads around to meet this need.

Increased demands of mobile users and talk of cloud computing architectures, however, are driving organizations to push these capabilities even further.  Enterprise’s want the ability to move applications closer to the users, move processing to dormant resources in international datacenters during off-peak hours and are increasingly interested in dynamically provisioning resources in third-party locations such as cloud providers and managed hosting providers.  This requires the ability to move massive virtual image files between distant locations in near real-time and the capability of dynamically rerouting application requests from one location to another—all with no impact to the users of those applications. These demands go way beyond the scope of server virtualization tools.

In January of 2009, F5 released BIG-IP v10.0. With it came a new functionality called iSessions. iSessions, included as part of the TMOS platform, provided the ability to create dynamic, optimized and secure tunnels between BIG-IP devices even ones in geographically disparate locations.  In August of that year, F5 announced and demoed the ability to use this functionality in conjunction with VMWare to provide uninterrupted access to applications as they were moved ‘live’ from one datacenter to  another.  With the release of BIG-IP v10.1 and the WAN Optimization Module, F5 continues to extend the capabilities and integration with VMWare to enable organizations to rapidly respond to changing application and business requirements by seamlessly migrating live applications across geographically dispersed data centers.

Now, F5 continues to lead the industry by publishing deployment guidance and test results that illustrate and validate the value of deploying F5^® and VMware solutions in concert to extend live migration capabilities across long distances for VMware vSphere^™ 4 environments in a secure, accelerated manner. 

“VMware VMotion^™ and Storage VMotion capabilities are significant differentiators for VMware vSphere^™. F5’s technology complements VMware solutions, making it possible to execute live workload migrations over greater distances. Combining VMware vSphere with BIG-IP^® solutions expands the many use cases possible, and adds to the value we’re able to offer customers.”- Parag Patel, Vice President, Alliances at VMware.

By integrating elements of F5’s application delivery solutions, Local Traffic Manager (LTM), Global Traffic Manager (GTM) and the new WAN Optimization Module (WOM), the F5 solution provides several benefits to the overall solution.  WOM provides the secure and optimized datacenter-to-datacenter connection that enables the workload to be migrated in a fraction of the time.  Once the workload has been transferred, GTM directs new user requests to the new location of the application. LTM, again utilizing the WOM services, enables existing user sessions to continue with minimal impact even though the application is now actually served from the other datacenter by routing user’s requests until they terminate.

According to recent ESG research, increasing server virtualization usage is the top overall IT priority facing organizations over the next 12 to 18 months. The mobility of virtual machines now provided by F5 and VMware will help companies distribute workloads across data centers, relocate applications closer to the end-user, and perform live data center migrations.” - Mark Bowker, Senior Analyst with Enterprise Strategy Group.

This solution enables customers of both F5 and VMware to see the benefits and possibilities of taking an integrated and holistic approach to their virtualization efforts and receive the full value of their investment.

Press Releases:

• F5 and VMware Solution Enables Cloud Connection for Live Application Mobility
• F5 and VMware Solution Yields 10x Improvement in Long Distance Live Migration Performance Using VMware VMotion

Additional Resources:

• Introducing: Long Distance VMotion with VMWare
• Deploying the BIG-IP System to Enable Long Distance VMotion with VMware vSphere
• Connecting to the Cloud with F5 BIG-IP Solutions and VMware VMotion
• F5 and VMware Enable Long Distance VMotion

VMware Information:

• VMware vSphere 4,
• VMware VMotion^™,
• VMware Storage VMotion
• VMware vCenter^™ APIs
• VMware vCenter Server

A complete guide to news and  media coverage for F5 Networks for the week of March 8th, 2010.

Notable Quotes:

Nathan Meyer, product manager at F5, said the benefits in upgrading systems during the deployment will be new systems that provide policy automation and key management. For example, Meyer said F5; a mid-level enterprise has over 100 different zones making the process of managing them a fairly daunting task.

"The task of managing each zone individually with individual keys and maintaining all the roll over periods would be a very daunting task," Meyer said. – Nathan Meyer, as quoted in Search Security

With each new incarnation of IT architecture and design, complexity and management seem to continually be a point of contention.  The advent of server virtualization and ‘VM sprawl’ has certainly highlighted the need for new ways to tackle the existing dilemma.  If managing a datacenter full of physical machines was difficult, managing the increasing number of virtual machines is another challenge all together.

In fact, one in four poll respondents admits to using the “wink and a
prayer” method of managing VMs and hypervisor hosts.(Hernick, 2008, p. 7)

This challenge is complicated even more when talk of virtualization turns to talk of dynamic provisioning of resources, cloud computing models which potentially push virtual instances outside the locally-owned datacenter and, now, the use of virtualized infrastructure components, virtual appliances and ‘infrastructure in a box’ application deployments. Without proper management practices and the systems/software to support them, this new world is quickly going to get out of hand and need more than just “a wink and a prayer”

F5 recently announce a new version of Enterprise Manager(tm) (EM); v2.0.  EM is a centralized management platform designed to give customers a consolidated, real-time view of their application delivery network (ADN) as well as the tools to manage the efficiency and performance of that network. This is an important differentiation.  EM provides two sets of tools to manage application delivery regardless of where the devices are physically located. 

The first set of tools is designed to manage the application delivery infrastructure itself.  This provides a simplified way to manage configuration, version control, policy enforcement and general maintenance of the application delivery controllers deployed throughout a customer’s infrastructure.  Now, tasks like ensuring updated SSL certificates across all devices serving the same application can be done quickly with little risk of error over traditional manual processes; and changes are implemented across the infrastructure at the same time.  Software version control and automated update facilities ensure that an organization can tell, at a moments notice, what the current state of their ADN is as well as seamless tools to correct issues if necessary.

The second set of tools provides a direct view into the health and efficiency of the ADN and the individual applications that are being delivered.  This allows organizations to better manage the deliver of their application by giving them the information they need. More than 160 customizable metrics capture current and historical data from your F5 BIG-IP devices. With total visibility into your application delivery infrastructure over time, you can analyze the data to troubleshoot performance and availability issues and provide better information for budgeting and capacity planning.

  As enterprises look to scale, consolidate, and automate network functions, solutions like F5’s Enterprise Manager continue to demonstrate attractive ROI for customers. Unified device management capabilities, comprehensive visibility into application delivery environments, and granular control over application resources should be primary considerations as organizations seek to minimize IT complexity or evaluate cloud and other resource-on-demand computing models. -- Zeus Kerravala, Senior Vice President of Enterprise Research at Yankee Group.

At the same time, F5 also announced the new 4000 series platform in support of the new version of EM.  The 4000 hardware platform allows customers with as few as eight devices to take advantage of Enterprise Manager’s sophisticated reporting, device monitoring, and performance management capabilities.

Together, the new capabilities of EM and the new hardware platform help F5 customers get a handle on managing the increasingly diverse and complex world of application delivery; even when those applications become virtualized and escape the confines of the local datacenter.

• Press Release
• Product Information:
• Enterprise Manager Product Page
• Enterprise Manager Data Sheet

References:
Hernick, J. (2008). The New Sprawl: ManagingVirtual Server Environment. Information Week. Retrieved March 13, 2010 from http://www.eginnovations.com/web/news/InformationWeek_VMSprawl.pdf

Founded in 1959, Computer Sciences Corporation is a leading  global IT services company. CSC provides innovative solutions for industry and government customers around the world by applying leading technologies and CSC's own advanced capabilities. These include systems design and integration; IT and business process outsourcing; applications software development; Web and application hosting; and management consulting.

In this podcast, we talk to Evan Loats, Remote Access Product Manager, Australia and Asia CSC, about the requirements of remote access users in today’s rapidly evolving mobile environment and the solution F5 Networks provides to their customers.

Podcast:
http://devcentral.f5.com/weblogs/interviews/archive/2010/01/22/talk-with-evan-loats-from-csc-about-remote-access.aspx

Product Information:
BIG-IP Edge Gateway: http://www.f5.com/pdf/products/big-ip-edge-gateway-overview.pdf
BIG-IP Access Policy Manager (APM): http://www.f5.com/pdf/products/big-ip-access-policy-manager-overview.pdf

Data growth continues unbounded.  This is easily demonstrated by comparing the 10MB disk available for the original IBM PC with the 320GB, 500GB and even 1TB drives that are often standard today.  If personal data has grown this tremendously you can only imagine the growth of enterprise data as more and more information is created, digitized and stored for longer periods.

Amount of information created will grow from 1.1EB in 2010 to ~50EB in 2020;  25 Quintillion Files – IDC Directions 2010 Key Note Presentation 

This staggering growth can cause significant impacts on CapEx as organizations struggle to keep up with the required amount of storage and even more so on OpEx in terms of backup, recovery and simply moving data from old, exhausted disk systems to new ones and the resultant client impact.

Growth of unstructured data continues to accelerate and data center administrators are tasked to optimize access, provide increased mobility, and deliver solid cost-containment benefits. Looking forward, vendors that can extend their solutions’ data management and virtualization capabilities through strong industry solution partnerships should rank highly on organizations’ wish lists.” - Dave Russell, Research Vice President at Gartner, Inc

F5’s ARX solutions provide the same kind of flexible, scalable and adaptable solution for storage systems that our BIG-IP solutions provide for applications.  By enabling file virtualization and separating the physical location of files from the way users and systems access them, ARX can provide tremendous benefit to customers dealing with the exponential growth in data.

By partnering with a recognized leader in storage like NetApp, F5 continues to provide real value that integrates with customer’s existing environments instead of simply providing point solutions that only address point problems.

The NetApp Alliance Partner Program brings like-minded solution providers together to provide cost-effective and validated solutions to our customers to help them simplify their data infrastructure. By joining the NetApp Alliance Partner Program, F5 will provide a complementary solution to NetApp storage that will help customers eliminate the many problems typically associated with data migration. Together, we’ll help customers boost productivity and increase the agility of their infrastructure. - Patrick Rogers, Vice President, Solutions Marketing at NetApp.

Press Release:
http://www.f5.com/news-press-events/press/2010/20100121.html

Solution Documentation:
http://www.f5.com/solutions/technology-alliances/infrastructure/netapp.html

Product Information:
http://www.f5.com/products/arx-series/

The evolution of network-based load balancers into today’s application delivery controller (ADC) is a perfect example of how today’s requirements are not necessarily tomorrow’s.  The idea of SSL termination, caching and intelligent compression or even layer-7 functionality simply didn’t exist when the first load-balancers came onto the market. Extensibility is the difference between a point product and a solution that provides value tomorrow and the next.  Extensibility, however, isn’t just bolted on—it has to be designed in and anticipated. 

This is perhaps even more relevant today as industry, vendors and customers alike continue to discuss cloud computing and the impacts to traditional architecture it is likely to have.  Not only must we contend with application delivery to an increasing number of diverse clients and multiple access paths, but now the application itself can move and change as it makes its way around the cloud—even if that cloud is built within your own datacenter(s).  This requires a whole new set of functions that enable organizations to control access to those applications, legal/regulatory requirements that may change depending on who accesses the application and from where, and the need to integrate WAN optimization into the whole orchestrated cloud process.  And, it sure would be nice to not have to buy more, new hardware and re-architect our networks to make this happen.

  F5 recently made a second wave of announcements concerning the capabilities of its BIG-IP v10.1 software release.  While the first announcement in November of last year concerned augmenting web-security capabilities and the specific needs of intelligently scaling mobile infrastructure, this announcement was almost entirely about extensibility, delivering added value and new unique solutions to the market.  Specifically, F5 announced new access control, WAN optimization and IP Geolocation services to the BIG-IP family of products.

  Access Policy Manager (APM)  provides a new set of tools to the application delivery process that simply didn’t exist before.  Now, customers can make access decisions to the application based on the same kind of security measures traditionally reserved for SSL VPN solutions—like enforcing encryption and client integrity checking to verify that the client adheres to the required security posture—and have unique requirements on a per application basis all on the same ADC.  More information on unified access and application delivery . . .

The WAN Optimization Module (WOM) continues to deliver advanced optimization features—right from the ADC.  Since the release of BIG-IP v10.0, F5 has enabled the TMOS base platform (included with all BIG-IP products) with iSessions, a way to create dynamic, secure and optimized tunnels between BIG-IP devices using symmetric adaptive compression, SSL, TCP optimization and bandwidth allocation.  With the release of v10.1, F5 has added Application Layer Acceleration and Data De-duplication services as

well with the addition of WOM.  Now customers with more demanding needs—like the real-time migration of virtual machines between remote datacenters have complete WAN optimization services at their fingertips in their existing ADC deployment.  More information on WAN application delivery services . . .

IP Geolocation Services provides F5 customers with new, intelligent application delivery options based on the highly accurate data provided by Quova.  This enables customers to use location data to accurately deliver applications to users—whether that means routing requests to the most appropriate datacenter or even denying users access to fulfill contractual or compliance mandates like broadcast restrictions or privacy laws.Since this information is part of TMOS, it is no longer restricted to only the Global Traffic Manager (GTM) product but can be used by all BIG-IP family products like the Application Security Manager (ASM) which can now log threats more accurately or provide users of the Local Traffic Manager (LTM) with detailed views into user distribution and behavior based on geographic origin. More information on IP Geolocation Services . . .

Put together, F5 continues to push the bounds of what is considered the normal features of Application Delivery Controllers, continues to provide rich, new services that help customers realize additional value and proves what extensible really means.

• Press Release 
• Access Policy Manager
• WAN Optimization Module
• F5/Quova Technology Alliance

Customer:           Confucius Institute Online
Vertical:              Education 
F5 Solution:        BIG-IP LTM, GTM, VIPRION ARX 
Partner:               Digital China
                             Sinogrid Information Technology

The majority of DNS infrastructures are vulnerable today. More and more customers are looking for a way, and allocating larger budgets, to protect and secure their DNS Infrastructures.  Recently there have been several high profile attacks on the DNS systems.  F5 and Infoblox together are uniquely positioned to solve these problems and secure the DNS infrastructures.  Every organization should be thinking about the security and scalability of their DNS infrastructure.

The combination of F5’s and Infoblox’s appliances provide enterprise customers an opportunity to build authoritative DNS infrastructure without giving up either global server load balancing or DNSSEC — it’s a no compromise solution.”

– Cricket Liu, Infoblox VP of Architecture and author of O’Reilly book DNS and BIND

F5 and Infoblox have joined forces to deliver the market’s only fully integrated and complete DNSSEC solution including high-performance DNS and global server load balancing functions, all supporting signed DNSSEC data. This provides customers a scalable, manageable, and secure DNS infrastructure that is equipped to withstand DNS attacks. The solution is a combination of Infoblox’s purpose-built appliances that deliver highly reliable, manageable and secure DNS services with built-in, automated DNSSEC features, and F5^® BIG-IP^® Global Traffic Manager^™ appliances optimized with hardware acceleration facilitating real-time signing of DNSSEC signature queries.

Together F5 and Infoblox provide a complete DNSSEC solution focused on:

• High Availability / Scalability
• Context-Aware
• Simplified DNS Management
• End-to-end DNSSEC

DNSSEC is still largely in its implementation stage for organizations outside the federal government, but the time is rapidly approaching when DNSSEC on top level domains will be complete.

Working methodically and carefully, and in collaboration with the U.S. Department of Commerce and ICANN, VeriSign anticipates that by the first quarter of 2011, DNSSEC implementation will be complete on the .edu, .net and .com Top Level Domains (TLDs). – CNN News

It’s time for organizations that have put off DNSSEC because of its complexity and the disruption caused by implementation to look into solutions and lay out an implementation strategy. A joint F5 and Infoblox solution can dramatically decrease the complexity and disruption implied by such implementations without negatively impacting existing global load balancing services that could interrupt disaster recovery and business continuity solutions.

Additional Resources:

• Joint F5 & Infoblox DNSSEC Webinar – March 10th
• F5 Infoblox DNS Integrated Architecture [PDF]
• DNS Security for web-based applications
• DNSstuff.com Drops Downtime from Minutes to Milliseconds with F5 BIG-IP Devices 
• It's DNSSEC Not DNSSUX
• External Resources
• 80% of Government Web sites miss DNS security deadline
• Twitter Has Been Hacked [Update: DNS Records Compromised]
• VeriSign Opens New Lab to Test Interoperability of Internet Solutions with DNSSEC
• Swiss Among World Leaders in Enabling DNSSEC
• First root server provides a DNSSEC-signed zone as of December 1st

“F5 is the market leader in application availability and performance, and also the market leader in data center redundancy solutions. The customer loved F5 for its ease of  management, ease of use, overall capabilities, and performance.”                                                  -- Jerry O’Brien, Enterprise Account Executive, FishNet Security

Customer:           Publically traded U.S. financial services institution

F5 Solution:        BIG-IP LTM & GTM

Partner:              FishNet Security

Customer Business Challenge

A publically traded U.S. financial services institution based in the Midwest recently came to FishNet Security with two problems. First, the company needed to achieve redundancy for its data centers so that in case of a catastrophic event, business would not be disrupted and customers at its 136 locations around the country would still be served. The company also needed to improve the availability and performance of all business applications running in its two data centers. The Cisco Content Services Switch that was in place was reaching the end of its support contract, and the customer was looking for a more robust solution.

In addition, the customer required that FishNet Security find one single vendor to solve the two challenges.

Customer Business Benefits

• Complete data center redundancy in case of outage or catastrophic event
• High-performance application delivery solution that the customer wasn’t getting from its existing equipment
• Enhanced information security
• Single vendor to provide comprehensive solution and support future business needs

Because we’re bound up in technology the term “rack space” immediately brings to mind rows of racks replete with servers humming and lights blinking merrily. But there’s another kind of rack that’s much quieter, has no lights, but still incurs similar costs associated with temperature control and circulation, requires maintenance and staff to manage, and that requires a whole lot of floor space.

That’s right, it’s your local library.

There are so many similarities between libraries and data centers – and in particular storage requirements – it’s amazing the analogy doesn’t crop up more in technical articles attempting to explain the benefits of cloud computing (off-site high density storage for aging books) and storage virtualization (tiering storage of volumes based on frequency of access).

Consider the cost differences between storing a volume in the library and storing it digitally in a data center, as referenced by this article on e-library economics:

Print volumes are, after all, voluminous — a property that implies a series of relatively pricey preservation costs. According to Courant and Nielsen, these work out to an average of $4.26 per book, per year when you take into account, maintenance, cleaning, electricity for temperature control, staffing, and circulation, as well as the considerable funds that go into building and renovating centrally located, open-stack facilities to house the volumes.

E-books are cheaper across the board — most notably in space and maintenance. Courant and Nielsen don’t get into precise modeling for e-book storage, but they note that the digital media repository Hathi Trust stores five million copies at $0.15 per volume, per year (that cost could rise to $0.40 for color volumes). Not only can e-book databases put many more books at scholars’ fingertips, but the medium seems intuitively suited for long-term storage: “While [print] books deteriorate with use, the reliability of e-books tends to be improved with use,” Courant and Nielsen write. (Though, as Henry and Spiro note, as technology evolves preservationists face a challenge in making sure e-books remain compatible with the hardware used to access them.)

“Where it is legally and functionally possible to make the move to electronic storage and use of the working copies of academic materials, there is substantial economic gain,” Courant and Nielsen add.

The automatic storage tiering capabilities of storage virtualization solutions makes it a perfect fit for such initiatives. As books age and are accessed less frequently, they can be moved to less expensive, slower storage systems that may be, for economic reasons, offsite. They’re still accessible, and can be moved back into priority tier storage upon being requested, but in the meantime they’re not taking up “rack space” in the physical library nor are they incurring the cost of being stored on expensive, fast storage arrays.

F5 ARX series provides the core storage tiering capabilities necessary to architect this type of solution. But just as important, perhaps, is the ability to create a secure, optimized tunnel over which files, such as digital books, can be transferred efficiently when first tier storage is local and secondary storage may be remote. F5 WAN Optimization Module (WOM) creates such a “tunnel” between sites, optimizing the exchange of data through network and application acceleration techniques including data de-duplication, caching, and compression. Tunnels are also secured using industry-standard SSL, which ensures safe transport of data to and from remote and central locations. F5 EDGE Gateway makes deployment at remote locations less costly, less disruptive, without sacrificing optimization and acceleration capabilities needed to assist in improving the timeliness and costs associated with movement of large data sets between locations.

By leveraging an F5 enabled architecture, you can make the most of your existing storage infrastructure wherever it may reside.

Related resources:

• SEGA – Simplifies File Management and Archiving System 
• Enterprise-Class File Virtualization 
• Disk May Be Cheap but Storage is Not
• Storage Tiering with ARX saves Architectural Firm More Than 5000 hours of User Productivity a Year
• Reason #1 That You Need File Virtualization
• Reason #2 That You Need File Virtualization
• Reason #3 That You Need File Virtualization
• Reason #4 That You Need File Virtualization
• Reason #5 That You Need File Virtualization
• RHWL and Storage Tiering [Case Study, PDF]
• External Resources
• What is File Virtualization – Storage Switzerland
• File Virtualization – a bridge to the Cloud – Storage Switzerland
• File Virtualization – Filevirtualization.com

A recent article asserts with its title that “95% of email is spam.” The article backs up its title with information gleaned from a recent study conducted by the European Network and Information Security Agency (ENISA) on the subject of spam:

The European Network and Information Security Agency (ENISA) released its new spam report which looks at spam budgets, impact of spam and spam management. The survey targeted email service providers of different types and sizes, and received replies from 100 respondents from 30 different countries, throughout the EU (26/27 EU Member States); and 80 million mailboxes managed. The survey analyses how e-mail service providers combat spam in their networks, and identifies the state of art in the fight against spam.

…

Less than 5% of all email traffic is delivered to mailboxes. This means the main bulk of mails, 95%, is spam. This is a very minor change, from 6%, in earlier ENISA reports.

The report also indicates that junk e-mail, i.e. SPAM, has a more significant impact on infrastructure costs and management than it does on bandwidth. Symantec also released a report last year indicating that 90% of all e-mail was junk. Much of those costs come from the hardware and software necessary to scan worthless e-mail for viruses and other malicious content. In order to keep performance at an acceptable level, it may require additional investments in more powerful hardware or more hardware running more copies of anti-virus scanning software, which incurs additional licensing costs from the operating system through the packaged software solution.

If an organization could reduce the amount of junk e-mail that must be processed by the infrastructure, it’s quite likely that the costs associated would also be reduced. Eliminating 60% of all junk e-mail from requiring corporate resources, then, should provide a significant reduction in the necessary hardware and software required to scan and assure the overall quality of the remaining e-mail.

The BIG-IP^® Message Security Module is a reputation-based, perimeter anti-spam solution integrated into the application delivery control network. The module leverages reputation data from Secure Computing's TrustedSource™ multi-identity reputation engine, which allows it to extend security for message applications to the edge of the corporate network and eliminate unwanted e-mail.

What message security does is check the reputation of the e-mail sender before the mail can enter the data center and rejects those messages that are identified as SPAM based on the behavior of the sender and associated domain. This prevents junk messages from being processed by the mail infrastructure, consuming valuable resources and taking up space on expensive storage systems due to retention policies. Reducing the volume of messages that must be processed by the infrastructure – especially anti-virus scanning systems – can significantly reduce the costs associated with mail infrastructure through decommissioning of the hardware and software necessary to support the extraordinarily high volume of SPAM processed by organizations every day.

Message security further enhances the security posture of e-mail by reducing the number of potential attack vectors through which phishing and malware can be delivered. Employee education is the foundation upon which e-mail security should be based, but eliminating the possibility by not delivering phishing and malware-laden e-mail to end-users enhances the overall success of mail security initiatives.

You can find more information about BIG-IP Message Security Module (MSM) here, and read about how F5 uses its own solutions to significantly reduce the volume of illegitimate messages processed by its own infrastructure in this case study.

Related resources:

• F5 Drops 60 Percent of Spam with BIG-IP Message Security Module 
• F5 MSM Case Study [PDF]
• American Internet Services Case Study | F5 Networks [PDF]
• External Resources
• 95% of E-mail is SPAM
• European Network and Information Security Agency (ENISA) 2009 E-mail Report [PDF]
• 90% of e-mail is spam, Symantec says [ComputerWorld]

The challenges associated with cloud computing and remote data center locations are strikingly similar, particularly with regards to performance and transfers of large files across bandwidth constrained, e.g. WAN, links.  Congestion, latency, and prioritization of other applications can slow down the transfer of data across data centers, whether traditional or cloud-based. Conversely, the utilization of Internet-connected WAN-speed links to provide both transfer of large data files and application service to myriad remote employees and users via mobile devices, roaming laptops, and dedicated desktops in remote offices, can cause performance challenges.

The challenges are only going to become more complex as more and more users, employees, and customers take advantage of advances in technology to effectively maintain an “always on” connection with the services you provide. This is more challenging than ever before because of the increasingly flexible manner in which applications are accessed across a variety of devices and locations, sometimes simultaneously.

IDC also reported in its new Marketplace Model and Forecast report that more than a quarter of the world's population, or 1.6 billion people, used the Internet in 2009 on a PC, mobile phone, video-game console, or other device. By 2013, that number is expected to rise to 2.2 billion.

-- InformationWeek, “1 Billion Mobile Internet Devices Seen By 2013”

What’s necessary to address the performance challenges of today’s volatile and mutli-faceted network environments is to be able to first identify who is accessing what application from where and what type of device so that the appropriate mechanisms to address the specific issues raised by a given combination of device, user, network, and application can be effectively utilized.

F5 addresses these challenges and provides the means by which organizations can apply optimization, acceleration, and security policies based on a variety of parameters. This allows organizations to specifically target devices, users, locations, applications or any combination thereof with the right set of solutions at the right time to mitigate performance and security challenges introduced by today’s increasingly mobile user-base and application deployment strategies. The solution to these problems does not, surprisingly, introduce more complexity into the data center. On the contrary, F5’s solution to these challenges actually simplifies the application delivery network by unifying many of these highly related functions onto a single, easily managed platform.

F5’s BIG-IP^® Edge Gateway^™ unifies secure SSL VPN access, dynamic access and optimization control, and application acceleration for remote users—regardless of the device or access network—on a single, scalable platform. BIG-IP Edge Gateway accelerates and optimizes applications, offering significant performance gains (up to eight times faster), and ensures that users are always seamlessly connected via F5’s new BIG-IP Edge Client^™ with smart connection technology. Smart connection technology’s unique persistence and access intelligence makes sure that as a user roams networks—wireless (802.11), cell (3G and LTE), or wired—they stay connected, and the integrity of their application session is protected. The client also includes best-in-class endpoint security and unique application and network acceleration.

Reducing the impact on transfer of large data files such as virtual machine images deployed into cloud computing environments or secondary data centers, BIG-IP WAN Optimization Module^™ (WOM^™) accelerates data transfer over the network by optimizing traffic and bandwidth through the use of adaptive compression and de-duplication technologies. WOM reduces bandwidth costs while increasing transfer rates up to 70 times over existing methods. This new BIG-IP product module ensures data is replicated quickly and reliably at speeds greater than 1 Gbps—faster and less-expensive than any alternative on the market today.

By leveraging both Edge Gateway and WOM, organizations can reduce bandwidth consumption and ensure no application or data transfer suffers to the benefit of others. Leveraging Edge Gateway organizations can target acceleration policies based on the unique characteristics of the endpoint, the network, and the application and ensure that IT resources are not wasted by being applied to applications and users that will not benefit from them.

You can find more information on Edge Gateway and WOM at F5’s corporate website.

Related resources:

• F5 BIG-IP v10.1 Advanced ADC Features [Slideshare Presentation]
• BIG-IP Version 10.1: An Integrated Application Delivery Architecture [Whitepaper, PDF]
• Geolocation and Application Delivery [Whitepaper, PDF]
• Unified Access and Optimization [Whitepaper, PDF]
• F5 Delivers Next-Generation Application Delivery Services Giving Enterprises More Control with Context-Aware Networking [Press release]
• BIG-IP v10.1 Now Available 
• Accelerating Your (Secure) Ride to the Cloud: Drive Smart(er) 
• Accelerating Your (Secure) Ride to the Cloud: Fasten Your Seatbelts
• Geolocation Gets More Granular with F5 BIG-IP and Quova 
• External articles:
  • F5 Adds SSL VPN to its Big-IP
  • F5 Reigns in Both Application Access and Remote Locations
  • F5 Intros BIG-IP Edge Gateway Solution to Offer Next Gen Remote Solution
  • F5 Networks Moves to Consolidate Services

Customer: SEGA

Vertical: Interactive Entertainment

F5 Solution: BIG-IP ARX

Business Challenge:

Over the past couple years, with the increase of high-quality videos and multimedia files SEGA faced growing data management challenges adding a significant burden to the file servers used by the company's nearly 4,000 employees. Weekly backup operations took 40 hours to complete and so SEGA began looking for ways to simplify the operation of its file servers and make its data management efforts more efficient.

F5 Solution:

Leveraging F5’s ARX's file virtualization solution to implement an automated storage tiering strategy, SEGA now has the flexibly to allocate storage resources, archive non-critical data on lower-cost, large capacity devices and free up primary storage systems for its frequently used data. This approach helped the company decrease the number of expensive primary storage systems required, and substantially reduced the time required for backup operations. By enabling SEGA to maximize the efficiency of its existing storage hardware, the company was also able to pool available storage resources from servers that had previously operated independently.

Customer Benefit:

• Reduce the Cost of Data Management and Backup
• Easily Manage Data without Impacting Users
• Conveniently Scale their Infrastructure

Related Resources:

• For more information about F5's ARX file virtualization solutions, please visit
  www.f5.com/products/arx-series/
• F5 Helps SEGA Reduce Data Management Costs by 66%
• F5's storage tiering solutions
• SEGA Corporation

Many IT departments are struggling with rapid increases in both traffic volumes and users. Simultaneously, organizations continue to leverage social media, mobile devices, and an expanding array of web services to achieve business goals. These demands result in many companies pushing the limits of their current application delivery infrastructures. Even the most capable of chassis based solutions may need a boost in performance and total capacity to support highly trafficked sites that are more on par with the demands placed upon the infrastructure of a service-provider class organization than a traditional enterprise. Service-providers, too, are feeling the pressure as more and more customers demand more services, faster, and more often from their mobile devices, seeking always-on data and voice communications with their friends, family, and coworkers regardless of where they might be.

Not to be forgotten, cloud providers are gaining customers and certainly looking for solid, flexible infrastructure to support not only their needs but their customers as well. An application delivery controller capable of integrating with automated provisioning systems and supporting a wide variety of application and network protocols simultaneously is a requirement for cloud providers looking to differentiate their offering from their competitors.

Organizations already relying on VIPRION will find relief from rapidly increasing demands on application delivery infrastructure with F5’s new Performance Blade 200 (PB200). The PB200 has dual quad-core processors that leverage F5’s Clustered Multiprocessing (CMP) technology and the blades work together form a virtual processing fabric utilizing all available CPU resources to meet users’ application delivery demands. Organizations outgrowing their existing application delivery controllers and looking to advance to the next level of availability and high-performance can take advantage of improved performance for VIPRION with PB200 hardware.

The new VIPRION system with PB200 hardware enables organizations to:

Microsoft Forefront Unified Access Gateway (UAG) is the evolution of Microsoft’s Intelligent Application Gateway (IAG 2007). Forefront UAG is a secure remote access solution focusing on applying application intelligence and granular access control across applications to enable easy management of access to enterprise applications by mobile and remote employees and partners. Forefront UAG provides centralized management and policy control across all users, devices, and network resources. 

Scaling solutions such as Forefront UAG is often challenging because of the unique requirements raised by such solutions, such as routing server generated connections in situations where the client has a pre-established tunnel to the correct server. This type of persistence of connections requires that the infrastructure solutions used to scale and provide high-availability for Forefront UAG be application-aware and capable of inspecting messages exchanged between the client and Forefront UAG to ensure existing connections are directed to the appropriate server. Without this ability the connections could be directed to the wrong server, i.e. a server without knowledge of the existing connection, and the connection would be dropped.

F5's BIG-IP Local Traffic Manager (LTM) can be used to provide scalability and high availability for Microsoft's Unified Access Gateway. When deployed on either side of the UAG servers, BIG-IP's load balancing capabilities can be leveraged to route both incoming and outgoing traffic through the most appropriate UAG server. BIG-IP LTM handles this traffic with its intelligent traffic engine, iRules, to track client-to-UAG server tunnels, and match server generated connections to the right UAG server.

It requires more than simple load balancing to properly scale and provide the high-availability necessary for Forefront UAG, including configurations to load balance both inbound and outbound connections. Such configurations can be complex, depending on the environment, and requires both network and application network layer skills. The need for persistence of server-generated connections requires a specific network-side scripting implementation, potentially extending the time to deployment.

To make the configuration of BIG-IP LTM simpler and easily reproduced, F5 has documented the process in a step-by-step deployment guide specifically for Forefront UAG. The deployment guide provides a complete methodology for configuring and deploying UAG in a highly-available, scalable environment, including leveraging IP-HTTPS for additional security.

The deployment guide is available now and can be downloaded here, through F5’s application-specific solution center.

Related resources:

• Microsoft Solutions Forum on DevCentral
• Deploying F5 with Microsoft Forefront Unified Access Gateway
• F5 Microsoft Solutions Center from F5
• Microsoft Forefront Unified Access Gateway Center from Microsoft
• External Articles
• Microsoft challenges IBM to Websphere duel • The Register
• F5 Blogs and Articles
• F5 PRO-enabled Management Pack: Managing Application Delivery in a Virtual Environment
• Microsoft Exchange 2010: HELO New Architecture

You’ve seen the press releases, blog posts, articles and tech tips and now the BIG-IP v10.1 code is available for download.*  From Optimizing Windows 7 & Server 2008R2 deployments to IP Geolocation to PCI Reporting to DNSSEC, BIG-IP v10.1 protects businesses from costly attacks, ensures compliance, streamlines management, and optimizes application performance.

BIG-IP v10.1 Download

BIG-IP Local Traffic Manager and TMOS version 10.1.0 (Release Notes)

*Registration and valid support contract may be required

            Customer: Hobsons

            Vertical: Education

            F5 Solution: BIG-IP GTM, LTM, WebAccelerator, iRules

Business Challenge:

Hobsons provides web-based SaaS solutions for colleges and universities for course planning, student retention, and career planning. Needing to push all of these services over the public internet while maintaining SLAs for their customers, Hobsons turned to F5 to ensure their mission-critical application delivery network was always available. Hobsons had to plan – while not over-provisioning – for traffic spikes, application migrations, and planned and un-planned maintenance periods by managing traffic direction and flow between legacy and applications for multiple customers and their users. “Our application delivery infrastructure is critical,” says Patrick McFadin, Director of Systems and Architecture at Hobsons.

F5 Solution: 

Using a powerful combination of F5 BIG-IP Global Traffic Manager (GTM), Local Traffic Manager (LTM), WebAccelerator, and iRules – F5’s application traffic management language – Hobsons is able to seamlessly maintain application availability by managing traffic from users to their entire application catalog across multiple data centers. LTM and iRules together are able to deliver application traffic to the appropriate application based on a list of situations, such as application availability, application response time, user connection type (mobile, LAN, broadband), and even a user’s authentication credentials. Once a user is directed to the appropriate application, WebAccelerator optimizes that user connection based on the above connection criteria, allowing Hobsons to always provide the best user experience possible for any application.

iRules in particular allowes Hobsons to not only manage user and application traffic in real-time but also to allow transparent application migration from legacy systems. iRules allows Hobsons’ users to access new applications without learning new complicated URLs or without changing hard-coded application settings and is able to re-writes web services calls between the legacy and new systems to help facilitate the transition, all without any burden or interaction on the end user.

Customer Benefit:

• Highly flexible application traffic management across multiple data centers
• Zero burden on end users
• Optimized user experience for all users
• Lower administrative and management costs
• Seamless migration of legacy applications

“Without the flexibility of the BIG-IP system and iRules, it wouldn’t have been possible to migrate users so seamlessly from our old data center to other ones. The new features of BIG-IP v10 made configuration tasks a lot easier and faster, which helped us meet our deadlines.” – Aftab Butt, Systems Architect, Hobsons

Case Study Details:

[PDF] http://www.f5.com/pdf/case-studies/hobsons-cs.pdf
[Audio] http://devcentral.f5.com/weblogs/interviews/archive/2009/12/07/delivering-saas-solutions-with-hobsons-patrick-mcfadin.aspx

Customer:       Northrop Grumman Corporation

Vertical:          Security

F5 Solution:     BIG-IP LTM, GTM, WebAccelerator, and multiple modules (see below)

The Army Knowledge Online/Defense Knowledge Online (AKO/DKO) Emerging Technologies Laboratory (ETL), established by Northrop Grumman Corporation, is a scaled replica of the AKO/DKO production environment that allows realistic testing, rapid prototyping

“ From the beginning of our implementation, key factors for success with the AKO/DKO ETL included an infrastructure comprised of the most robust solutions for application delivery,” said Skip Magness, AKO/DKO program manager at Northrop Grumman. “The sheer volume of traffic that the AKO/DKO portal generates required solutions that enhance security, manage and redirect traffic, reduce server loads, and provide for disaster recovery if necessary. With the continuing growth of service portal traffic, we also required scalable solutions to meet the needs of the users. We are seeing excellent results with the solutions we’ve implemented.”

To meet the diverse network and application networking needs, Northrop Grumman selected a variety of F5 technologies upon which to build their emerging technologies lab:

• BIG-IP Local Traffic Manager™ (LTM™) to load balance, secure, and optimize application traffic
• BIG-IP Global Traffic Manager™ (GTM™) for WAN high availability with seamless disaster recovery and routing
• BIG-IP WebAccelerator™ to reduce server processor utilization and bandwidth usage
• BIG-IP IPv6 Gateway™ module provides translation and load balancing between v4 and v6 networks, and directs traffic across mixed IPv6 and IPv4 devices
• BIG-IP Intelligent Compression™ module reduces the volume of traffic sent to users
• BIG-IP Fast Cache™ module prevents servers from repeatedly handling requests for identical content
• BIG-IP Advanced Client Authentication™ module eliminates the need to manage and enforce authentication individually across applications

Related Resources:

• F5 Networks Provides Key Infrastructure for Northrop Grumman's AKO/DKO Emerging Technologies Laboratory [Press Release ]
• Application Delivery & Optimization for Remote Users
• Innovative Web Protection and Compliance

File and data storage continues to be a challenge for IT departments trying to manage the ever increasing growth of information.  Users are always creating new data on a daily basis and regulatory requirements demand the preservation of information for longer periods of time.  Capacity demands are growing yet adding new storage devices can increase complexity and costs.  Many storage infrastructures consist of a mix of vendors that do not always work seamlessly together and users are statically mapped to physical file storage resources which can be disrupted when new devices are introduced or when files are moved.  Updates can cause downtime due to the manual nature of static configurations.  Finally, many companies overprovision storage capacity to meet the growing demands in the enterprise and costs can escalate quickly, especially for mid-size organizations. 

This week, F5 announced the new ARX Solution Portfolio, Delivering Enterprise-Class File Virtualization to Mid-Range Storage Environments on the ARX2000. 

‘Designed especially to meet the scalability and performance needs of mid-to-large size storage environments, the ARX2000 enables these organizations to more effectively address rapid data growth while reducing the cost and complexity of data management.’ 

File virtualization creates a logical, virtual representation of your storage environment, which eliminates the rigid mapping between clients and physical file location.  This virtualized environment enables simple logical access to physical file systems without introducing a new file system.  ARX file virtualization solutions act as a proxy to federate the file systems that are already in place on the underlying storage layer enabling IT organizations to automate and enforce data management policies in real time.  The ARX Data Management Operating System v5.1 enhanced support for Microsoft/CIFS storage environments which enables greater scalability for tier 2 deduplication platforms.  With the ARX2000, customers now have several ARX platforms to choose from, depending on their needs:

• ARX500 and ARX1000 for small to mid-size storage environments
• ARX2000 for mid-size environments: twice the virtualization scale & connectivity as the ARX1000
• ARX4000 for large environments

George Crump, Lead Analyst at Storage Switzerland says, "Getting the right data on the right tier at the right time has just been too user intensive. Administrators need a simpler and more effective method to move data seamlessly between tiers of storage, especially in the resource-constrained medium sized data center. File virtualization products like the ARX2000 should be a top consideration for these storage managers looking to realize the full ROI potential of storage tiering."

Resources:

• Intelligent File Virtualization Whitepaper (pdf) and Audio Version
• Reason #1 That You Need File Virtualization
• File Virtualization… The short primer
• Persistently Different - File Virtualization
• Disk May Be Cheap but Storage is Not
• How RHWL mitigated storage challenges with F5's ARX solution (Video)

The Domain Name System (DNS) was created in 1983 to enable humans to identify all the computers, services, and resources connected to the Internet by name.  Back then, security was not included in the original DNS design since at the time scalability, rather than malicious behavior, was the primary concern. Many feel that securing DNS would go a long way to securing the Internet at large.  Just this year, the main DNS registrar in Puerto Rico was hacked by a DNS attack in which local versions of Microsoft, Google, Yahoo, PayPal and others were re-directed to defaced or blank sites.  Also in 2009, one of Brazil’s largest banks got hit and a redirect took unsuspecting users to a malicious site which attempted to install malware and steal passwords.

With the release of BIG-IP v10.1, F5’s BIG-IP Global Traffic Manager (GTM) can provide real-time DNS Security with the DNSSEC add-on feature, protecting organizations from a host of DNS attacks.  DNSSEC ensures that the answer you receive when asking for name resolution comes from a trusted name server. Since DNSSEC is still far from being globally deployed and many resolvers either haven’t been updated or don’t support DNSSEC, implementing the BIG‑IP GTM DNSSEC feature can greatly enhance your DNS security right away. It can help you comply with federal DNSSEC mandates and help protect your valuable domain name and web properties from rogue servers sending invalid responses.

F5’s unique, patent-pending solution to the GSLB DNSSEC problem addresses DNSSEC by signing answers at the time the GTM device decides what the DNS response should be.  This is a real-time DNSSEC solution, and, with it, F5 is the only GSLB provider to have a true DNSSEC solution that works. While others have proposed a system in which every possible response is pre-signed, most have concluded that this isn’t a feasible approach.

From the press release:

Key DNSSEC features for BIG-IP GTM enable organizations to:

• Meet compliance mandates for DNSSEC by the 2009 federal deadline
• Sign DNS responses in real time and provide the means to deploy DNSSEC quickly and easily in an existing environment
• Determine accurately where a user is based on their IP address
• Provide the capability to sign the DNS responses to protect against rogue DNS servers
• Ensure that end-users receive correct web responses

The combination of BIG‑IP Local Traffic Manager + BIG‑IP GTM + DNSSEC on one box provides a drop-in DNSSEC solution for any existing DNS deployment, instantly giving you greater control and security over your DNS infrastructure while meeting U.S. Government mandates for DNSSEC compliance.  Rather than ripping and replacing your current DNS infrastructure, you can simply drop BIG‑IP GTM in front of your existing DNS servers and reduce your management costs with implementation and maintenance all on the same appliance.

Resources

Reference: Portions of this entry are excerpted from: DNSSEC: The Antidote to DNS Cache Poisoning and other DNS Attacks - White paper

Business Challenge:

As RHWL grew and started to work on more and more projects, it began to face a series of challenges in its UK office. “We were continually running out of storage space on our SAN solution,” says Dave Allerton, IT Director at RHWL. “As a result, we had to do emergency archives of data that, ideally, we would have preferred to have kept online.” Soon, these “emergencies” became an almost weekly event. “They started to take up a lot of time in the IT department and also inconvenienced employees,” explains Allerton. “Taking files offline required interacting with the architects to ask them which folders we could offload. Then, just as we finished offloading and freeing capacity, we would invariably need to refer to plans and documents that had just been archived, and then the IT team would have to reverse the process. This cost the business a lot of money in terms of wasted IT time, wasted architects’ time, and possible delays in design projects.”

Customer Benefits:

• More than 5000 man-hours of user productivity gained per year
• Optimized use of existing storage capacity, eliminating the need to purchase additional tier 1 storage
• Backup time reduced from 14 hours to 1.5 hours
• Reduced tape and offsite storage costs

Case study details:

[PDF] http://www.f5.com/pdf/case-studies/rhwl-architects-cs.pdf

 

For the past few years, F5’s VIPRION Application Delivery Controller chassis-based hardware platform has led the ADC market in application delivery performance  and scale. As a bladed system, VIPRION scales linearly to meet customer need for ADC resources: As application and network demands increase, VIPRION can grow and scale to meet those needs without adding additional new hardware platforms, all the way up to 1 Million layer 4 connections per second.

Under the hood VIPRION’s scale and massive application delivery performance is achieved using clustered multiprocessing – a technology that creates a virtualized application processing platform which can consume new hardware resources as needed. When performance needs exceed the resources available to the virtual platform a new blade can be added and those resources are consumed in real time by the virtual platform and made available to the application delivery traffic.

One of the challenges with the level of performance introduced by VIPRION has been how to adequately test these new-to-market performance benchmarks under real application load, and to test them in an aggregate manner. Simply stated, no tool existed to push VIPRION to it’s limits when using clustered multiprocessing and all blades as one unified and virtualized platform. 

On November 17th, 2009, F5 and Ixia announced the results of performance testing against a fully loaded and configured VIPRION chassis. The goal of this testing was to validate the aggregate performance numbers of a VIPRION chassis using real-world application traffic. Ixia has a long-standing reputation of developing tools to push F5 gear to their maximum in order to provide our customers with expected real-world results, and that trend continues today with these new clustered multiprocessing tests. Using a suite of Ixia products – IxLoad, XM12 platforms, and Acceleron-NP load modules – this new testing shows that VIPRION leads the ADC market in both network and application delivery performance.

Many vendors will release testing data based on theoretical numbers in lab environments using perfectly modeled network and application conditions, however real-world networks and application deployments rarely have the benefit of being pristine. Most application deployments include a mix of good and not-so-good network conditions, such as high-latency WAN links or other devices on the LAN causing network congestion. Working with Ixia, F5 has verified and validated that using conditions normally encountered on public and private networks, VIPRION can still manage application delivery across those networks to provide unprecedented performance numbers. Key VIPRION metrics from the Ixia testing include 1 Million layer 4 connections per second and 155,000 SSL transactions per second, performance unequaled in the ADC market.

Related Resources:

• More information on F5 VIPRION
• More information on the Ixia line of performance testing products
• Clustered Multiprocessing: Changing the Rules of the Performance Game (whitepaper)
• The ROI of Application Delivery Controllers in Traditional and Virtualized Environments (whitepaper)

The workforce continues to evolve and is more mobile than ever before. They are using an ever growing list of devices with access to more corporate information arriving from many different networks.  Companies are also opening access to internal systems to partners, contractors, guests and other users along with their non-IT trusted devices. While this trend helps the mobile force be productive, it can also introduce threats to the network. Traditionally, internal access to tele-workers has been granted via VPN connections, this too can be cumbersome for organizations trying to manage the multitude of devices requesting access from various bandwidth & types of networks.

Last month Microsoft announced the release of Windows 7 and with that, enabling clients with Direct Access and BranchCache, new features announced as part of Microsoft’s Windows Server 2008 R2 release.  Earlier this week, F5 Networks announced ‘new Application Ready Solutions for Microsoft Windows 7 and Windows Server 2008 R2,’ providing support for Microsoft's recently released DirectAccess and BranchCache technologies.

Direct Access provides users with the same experience working remotely as they would have when working in the office. With DirectAccess, remote users can access corporate file shares, Web sites, and applications without connecting to a virtual private network (VPN).  F5’s BIG-IP Local Traffic Manager now provides scalability and high availability for Microsoft’s DirectAccess deployments so customers can enjoy the same benefits of other ADC-optimized applications.  BIG-IP’s load balancing capabilities are designed to route both incoming and outgoing traffic through the most appropriate path to ensure that users get either the most available server (for a new connection) or the most appropriate server (for established ‘sticky’ sessions) for that request.  A nice solution overview, with illustration and iRule, can be found on F5’s DevCentral.

For BranchCache, we announced BIG-IP’s ability to both offload CPU intensive functions and optimize Windows 7 content requests even when a Windows Server 2008 R2 upgrade plan is still in the initial stages:

F5's support for BranchCache increases server capacity by offloading intensive computations used to alias web content requests from Windows 7 clients at branch locations.  This solution also functions as a transition technology to bring the benefits of BranchCache to all Windows 7 client users where web servers have not yet been upgraded to Windows Server 2008 R2.  - F5 Press Release.

Addressing the needs of the mobile worker and enabling IT with the tools to deliver, monitor and scale any application request from any user from any location.  IT Agility, Your Way.

Related Resources

• More information on F5 BIG-IP v10.1
• F5 Solutions Optimize Microsoft Windows 7 and Windows Server 2008 R2 Deployments
• v10.1 - BIG-IP and Microsoft DirectAccess
• pyControl Apps #1 - BIG-IP Routing Table
• Why should I care about Big-IP 10.1?
• F5 Updates Microsoft Exchange 2010 Solution Portfolio
• v10.1 - Configuring GTM's DNS Security Extensions
• F5 Highlights New Security Features In BIG-IP 10.1
• F5 bolsters BIG-IP security features