Wednesday, December 13, 2006
by Nathan Bragaw
We just entered into a strategic partnership with Network Instruments. Their GigaStor appliance captures and stores all the packets traveling networks for historical packet-level analysis. Our SuperAgent identifies the source of performance problems, application, server, or network, and isolates when and where they are degraded. Together, the products can monitor large networks, isolate performance issues, and provide packet-level analysis specific to the problem.
For example, this allows an engineer to identify a network that is showing slow web performance. The engineer can then drill down to a packet capture that includes network traffic 30 minutes before and 30 minutes after the event. This capability is changing the troubleshooting capabilities by replaying network traffic prior to the issue occurring. It's like having a TiVO for your network.
Engineers get an alert from SuperAgent, identifying the time and location of a network problem. Then you go into GigaStor and sort through the traffic to identify the root cause of the problem. We’re pretty psyched about GigaStor’s ability to reassemble packet streams to recreate e-mails, web pages, IM sessions and VoIP calls.
We call it Retrospective Network Analysis, or RNA for short.
We should have the products completely integrated sometime in the next six months or so. We’ve already started selling Network Instruments’ GigaStor products through our sales team and our Web site.
Tuesday, December 12, 2006
By Steven Maercklein and Zack Belcher
While we haven’t had a chance to play with Vista yet, (both of us are on the road all the time and we don’t have access to a lab,) we have been doing a lot of research on the new TCP/IP features of Vista and Longhorn.
We spent some time poring over a document from Microsoft’s Research Asia (the guys who designed the Compound TCP/IP [CTCP] algorithm) and it details how the algorithm works. It doesn’t detail how Microsoft has implemented the algorithm – there are alpha, beta, and gamma values that are tweakable, but it doesn’t go into how Microsoft has tweaked them. It looks like this was a document just to prove that CTCP was feasible and ready to be included in Vista.
It does teach you about the behavior of it and what to expect when seen in action, and it makes reference to the current technologies for congestion avoidance that CTCP is based on. However, there are several features of Vista of concern to those involved in network performance.
(Continued...)
Continue reading "Vista's TCP/IP Promises and Perils" »
Thursday, December 07, 2006
by Ted Romer
I would say, compared to Windows XP’s TCP/IP stack, the Vista stack has been redesigned from the ground up, instead of merely patched. The result is that there are a number of features in the new stack, (which Microsoft calls “The Next Generation TCP/IP Stack”), which are a bit exciting. (Windows Server Longhorn also runs the same stack as Vista.)
There’s been some pretty interesting changes with Receive Window Auto-Tuning and Compound TCP, which provide more aggressive scaling for the network window. Windows scaling is now enabled by default, and is also configured automatically.
If we think about a TCP conversation right now, a sender is given a window as to how much they can send without acknowledgement. The service slowly ramps up in a linear fashion, and if something happens (like a packet loss) we go back to a slow start and slowly ramp up again.
(Continued...)
Continue reading "Vista's "Next Generation" TCP/IP Stack and the Enterprise" »
Monday, December 04, 2006
by Chandra Hosek
Dr. Cathy Fulton, NetQoS’s chief technology officer, will be at the Computer Measurement Group’s (CMG) 32nd annual conference in Reno, Nev., on Thursday, discussing the trials and tribulations of achieving and maintaining optimal application performance across enterprise wide area networks.
In a session entitled “Best Laid Plans: Enterprise Network Performance Case Studies and Lessons Learned,” Cathy Fulton will share real-world case studies that demonstrate how well-intentioned network and systems engineering efforts can sometimes produce unexpected results - and how to avoid these mistakes - drawn from her years as a leading network engineering consultant and her in-the-trenches experience working in large enterprise network environments. Examples will include the results of implementing caching devices to improve application response times for remote users, improper application of QoS techniques, active agent monitoring software running amuck, and others.
NetQoS will also present a vendor training session at the CMG conference, from Dr. Steve Fulton, our director of product management, called “Why Utilization is not a Proxy for Performance.” Many IT professionals monitor utilization as a proxy for network performance, but modern techniques are proving this is not the best approach, especially with the advent of WAN optimization technologies. The most meaningful performance metric is end user response time captured by measuring real user transactions which, when combined with traffic flow data and traditional device statistics, enable network managers to optimize performance and accurately assess infrastructure needs. In his presentation, Steve will provide best practice examples that illustrate this new approach.
Ed Tittel, contributor to Network Performance Daily, has provided a rundown of events from the CMG 2006 conference at NetPerformance.com.
Chandra Hosek is Public Relations Manager at NetQoS.
Thursday, November 30, 2006
by Joel Trammell
Microsoft released Windows Vista (along with the new Office suite and Exchange server) to business customers today.
From a networking perspective, there have been several changes in the networking stack in the move from Windows XP to Windows Vista. We’re doing a fair amount of work, because we understand what Microsoft says they did, and we’re doing a fair amount of work to find out if that’s the way things actually work in reality. There’s testing that needs to be done, once network engineers understand the new Vista network stack, changes in the way that network engineers need to think.
For a networking person, I think Vista’s a pretty big deal, whereas some of the other Windows upgrades, maybe from a pure networking perspective, weren’t. You’re going to see some changes because of it. We'll have some details about how networking is different in Vista on this blog soon.
Joel Trammell is the CEO of NetQoS.
Monday, November 27, 2006
by Cathy Fulton
For any technical mistake, there’s usually a technical solution. Losing time while a problem persists is embarrassing, but you can recover from such delays. Much bigger problems occur when you try to change the workflow of the people who use the software. That’s when IT projects become difficult.
It’s extraordinarily difficult to get people to change the way they’ve been doing things. If you deploy an expensive piece of software in the expectation that people will automatically change the way that they normally do their jobs as a consequence of the switchover, you will greatly reduce your chances of success.
Continue reading "Perspectives on What's Missing in the Field of Network Management" »
Wednesday, November 22, 2006
by Carol Schiraldi
I’m not a gamer myself, but the recent launch of the PS3 and Wii consoles (podcast from Gamingbits available here) makes this a good time to discuss game programming and the idea of performance in the enterprise.
Austin is a big center for video game development – NCSoft, Midway Studios-Austin and Retro Studios are all located here – and I know many game programmers. I think that game programmers are some of the best programmers out there, and a lot of performance enhancements are coming from gaming.
If you code for gaming, there’s a different mindset. Emphasis on performance is included (and essential) by design.
Continue reading "Performance-Driven: Why enterprise developers (generally) use Java and game programmers (generally) use C++" »
