In 2003 Netli created their Application Delivery Network (ADN). This is considered by many to be the earliest "cloud-based WAN optimization" service. Now, "The Cloud" is a household word, and is being used as a platform for many business-critical services.
In the performance management arena, new tools and approaches have been and are being developed to address this rapidly-growing market. Netli has since become Akamai, and other cloud providers have appeared on the scene including Internap, Cloud Leverage, Virtela, and Aryaka.
One could rightly draw the conclusion that there is a solid presence for WAN-based traffic optimization, that it is becoming more and more mature, and that it is a natural extension to virtualize your key business applications. Why then wouldn't you virtualize application performance management as well? Tools are available that measure performance of virtualized business services, and now tools such as Ixia's IxVM suite place the tool frameworks themselves in the virtual environment. The advantages include better integration for verification of policy enforcement, VMAN trunking performance, port mirroring, and more.
Here is the question I would like to pose to the enterprise CIO: If you are investing in the migration of your crucial business applications to the cloud, then how do you best insure a successful outcome? Testing of key business applications, the infrastructure that they run on, and the devices that connect that infrastructure is a mature science. Now, we need to increase that sphere to include testing in the cloud as well!
I just finished listening-in on an excellent webinar put on by Shunra entitled "5 mistakes to avoid in data center relocations (DCRs)". I highly recommend it.
I listened-in because I always like to get another perspective from folks who are demonstrably smart at solving those problems. Shunra, by the way, is in the application performance engineering business, and produces many good and mature products that I consider extremely complimentary to Ixia's in solving similar issues in our common market.
The webinar made me recall the importance of cost avoidance in the enterprise, bringing to the surface my own strong feelings on the topic and its importance in making critical investment and deployment decisions.
Where application folks are concerned, cost avoidance means deploying applications only after we are quite sure of that application’s performance and resiliency under load. Where network folks are concerned, it means first insuring that the underlying IP transport will handle the applications and services that are clearly "business critical". Back in my LoadRunner days, my most memorable observation was the proven and documented savings a customer gained through cost avoidance. Today at Ixia, my goal is to take this same message to enterprise IT professionals who are looking to mitigate the risk of deploying systems and services that support key business applications.
While working for PowerTest in 2008, one of our customers, a “big three carrier”, showed me an internal report. It contained hard figures of savings through cost avoidance that were gathered after comparing downtime-related costs in a before-and-after comparison. The result was an astounding $1.4 million dollars saved over the first year by designing/implementing systems only after clearly mapping out the limits of performance and actually measuring, in a repeatable way, resiliency.
Ixia has similar success stories that are ongoing, including one case that focuses on end-to-end gateway testing, billing and charging mechanism validation, and wireless security testing – all for a production network that is literally the bread and butter if the organization. The cost savings from avoiding outages and rollbacks exceed the example given above.
The take-away here is that a well-running business infrastructure helps you deliver uninterrupted service, and deliver those services under budget. If you use sound practices for pre-deployment testing, you will understand a) where the acceptable performance limits lie and b) how to best stay within those limits.
Improved firewall support is now available for customers who want to use IxChariot in a strict firewall configuration, when specific source ports for all management connections must be set. This will entail a simple Console configuration. Endpoint machines need not be configured for this requirement.
Quality of service (QoS) validation has always been a strong force behind IxChariot's popularity. Recent additions to the QoS capability list for IxChariot include qWave QoS values for Microsoft Windows 7 and Microsoft Windows Server 2008 R2 platforms. In addition, the IxChariot Performance Endpoint for Linux now supports the Layer 2 Priority QoS field or CoS (class of service) by setting the Priority field of the VLAN as defined by IEEE 802.1p specification. The user configuration for Linux to support L2 priority QoS is straightforward. Give us a call for details!
Ixia recently held an online webinar entitled, "Testing the Cloud: Optimizing Virtualized Data Center Performance." The webinar featured two speakers that focused on the technology as well as how Ixia can address it with several products. The webinar also featured a live Q&A session that was interesting and informative. If you missed it, you can still catch the videos at the Ixia Webinar website.
Keep in mind that Ixia hosts live webinars once or twice a month!
Intel recently posted benchmarks of an 8-core Nehalem-based server with two dual-port NICs based on their next-gen 82599 10Gbps Ethernet chipset. Using IxChariot, Intel was able to demonstrate more than 50 Gbps of throughput bidirectionally on a single server. Intel's notes on the test setup... Performance result of a bandwidth intensive network benchmark (IxChariot). Network throughput was measured on 64KB I/O size transfers between the test system and multiple network targets. Intel pre-production system with two Quad-Core Intel® Xeon® processor 5500 series CPUs (2.93 GHz), 12 GB memory (6 x2GB DDR3 - 1066MHz) vs. Intel Production system with two Intel® Xeon® processors X5365 (3.0GHz, 1333MHz FSB), 8 GB memory (8 x 1 GB DDR 2 - 667). Windows Server 2008, stock unmodified installation. 
[External Link] Need to Get the Most out of 10 Gigabit? Use Nehalem.
Samara Lynn of ChannelWeb posted an article this week on products that boost network performance and help you get the most out of your infrastructure investment. That's exactly how customers are using IxChariot every day - making sure that SLA's are met, QoS policies are handled properly by every device in the network and performance is optimized for network conditions. For more information on tuning your network with IxChariot, or to share your successes with other Chariot users, see this thread on our TestTalk Chariot forum.
Originally posted April 6, 2009
Ixia has posted a new test plan on virtualization performance to IxChariot.com. This test plan explains how to use IxChariot™ to validate and optimize a virtualized data center. IxChariot™ is unique as it can generate and analyze application traffic between many endpoints and the VMs answering questions such as:
What is the optimum VMs to processor ratio for peak network performance?
How do multiple NICs impact the performance of virtualized servers?
What is the effect of latency introduced by virtualization on application performance?
Ixia is also hosting a Webinar called 'Assessing Network Performance in Virtualized Environments' on April 23rd at 10.00 am Pacific Time. During the webinar we will discuss network-related challenges when transitioning to a virtualized data center and how IxChariot™ can help you pinpoint those problems and find solutions.
We'd like to hear what you think of the test plan. Please post a comment here.
Originally posted on April 6, 2009 Are you ready for Microsoft Windows® 7? Let IxChariot help you prepare today! Responding to popular demand, Ixia’s engineering team has built new IxChariot™ endpoints for the public beta of Microsoft’s upcoming Windows 7 OS. This endpoint has been fully tested on both 32-bit and 64-bit versions of Windows 7 Beta build 7000, and the new endpoints are now available in the IxChariot™ Endpoint Library. The Windows 7 endpoints are completely compatible with all your existing IxChariot™ endpoints, so you can transition your testing seamlessly to the new OS. This allows you to quickly run functional tests, where you can conduct network performance comparisons between existing versions of Windows and Windows 7. If there are any differences, you’ll find them right away. Bottom line: IxChariot is ready to help you perform tests on the forthcoming Windows 7 Operating System. When Windows 7 hits the streets, you’ll be ready!
Originally published on February 2, 2009
Ixia just held a conference in San Diego with over 300 attendees from around the world. The participants were Ixia partners, systems engineers, trainers and network service professionals who use IxChariot.
During the conference we discussed the 2009 roadmap for IxChariot and the ways network engineers use IxChariot for different applications. One of the most interesting new applications was assessing the performance of datacenter virtualization. Server virtualization is one of the hottest trends in the IT industry. IT managers want to know: What impact does virtualization have on network performance?
It turns out that Steve Broadhead of Broadband-Testing Labs in the UK has been using IxChariot to answer that question. He recently ran a number of tests using two servers with 10GbE NICs running VMware and XEN virtual environments. His objective was to compare the performance of different 10GbE NICs from SolarFlare and Intel. Using IxChariot he was able to run up to 99% of line rate 10GbE throughput with up to 16 virtual machines per physical server.
Read Steve's full report here.
Let us know how you use IxChariot to test new networks.