Results of our WAN acceleration product feedback are in, and here is a quick summary of what we heard. For starters, if this were the application delivery system version of American Idol, Riverbed’s Steelhead would win.
Read more
We’ve received a lively response from many of you about experiences with your WAN acceleration vendors, and starting next week we will summarize what we heard. Before we tally up the responses though, folks have one last chance to weigh in. As reflected in the comment count below, thus far we’ve seen enormous variation in the relative participation and passion of different vendor’s customers. Let us know what you think, passionate or not. Although we don’t yet want to draw conclusions, feeble response may indicate lack of customer enthusiasm.
Read more
For the final installment in our "ADS Customers Speak" series we ask Silver Peak's customers to share their experiences with the NX product line. Arriving on the distributed application delivery system (ADS) scene a mere four years ago, Silver Peak is a relative market newcomer. During its short life, however, Silver Peak has hatched an impressive lineup of appliances ranging from its 'papa bear' NX-9000 that it claims is the largest WAN acceleration appliance on the market, to the 'baby bear' NX-2000 at home in small branch offices.
Read more
This week it is Expand Networks' Compass customers’ turn to share their experiences with the Network World community. A long-established player in the distributed application delivery system (ADS) market, Expand got its start in Israel ten years ago, and currently boasts more than 3,000 customers with over 36,000 units installed.
Read more
At the first ever O'Reilly Velocity conference in San Francisco yesterday we witnessed the debut of three new tools to measure website performance - KITE, Jiffy and CloudStatus. Not only are these tools very cool and useful, they are also free!
All about making websites faster and better, the conference was packed with more than 600 performance jocks - premiere among them Google's Steve Souders, who as Yahoo's performance guru developed a bag of tricks to make Yahoo speedy. Souders created YSlow, a Firefox add-on that analyzes web page performance, and wrote the book "High Performance Web Sites". Along the way he convinced O'Reilly to create the Velocity conference which he co-chairs.
Read more
Acceleration vendor marketing departments push out a huge body of misinformation every day. You have read the headlines: "We speed up your application response time 400X!" What does that actually mean? What is the math behind a 400X change? Here's the answer.
Take an example in which the base (pre-acceleration) response time for an application task (the time from when the user hits enter to when the screen refreshes) is 8 seconds. When the acceleration technique is applied to the application traffic flow, the same task now takes 2 seconds. The 6 second difference is a very respectable improvement which many users will notice.
Read more
Australian upstart Exinda Networks is in this week’s “ADS Customers Speak” spotlight. While leading US vendors like Riverbed, Cisco, Juniper, Packeteer and Blue Coat focus on high-end, high-performance application delivery solutions, this strappy little company from down under is racing after the middle market—which may well be larger overall than the top-of-the-line market. This week we ask Exinda’s customers’ to share their experiences with the Exinda x800 product.
Read more
Today we ask you to fill out our survey about application performance management best practices. We want you to tell us what you're up to, and how things are going with your application performance management efforts. We even bribe you to take part with the chance to win an Apple iPod Touch 16 GB.
We will write up the survey results in a Network World feature article and we will share additional details of what we learn on this blog. Go to the online survey by clicking here.
Among the applications most likely impaired by file server centralization are Word, PowerPoint, and Excel, which use Microsoft's Common Internet File System (CIFS) file sharing protocol, as well as Outlook which uses Microsoft's Messaging Applications Programming Interface (MAPI) email protocol for Exchange. CIFS and MAPI-based applications degrade badly as distance to users increases, and here's why.
The problem is that CIFS and MAPI are just plain poor network citizens. They were designed to run over local area networks, where the performance price for application protocol "chattiness" is negligible. Unfortunately, when run over a WAN, such chattiness (application turns) exacts a heavy toll.
Read more
This week in our "ADS Customers Speak" series we ask Blue Coat Systems customers to share their experiences with Blue Coat's Proxy SG product. We've gotten some great and in many cases passionate feedback from customers of other vendors, and now it's Blue Coat's turn.
Read more
Behemoths and little guys do a better job of application performance management than mid-sized enterprises. That's what we found in our last benchmarking survey of enterprise application performance management practices. Enterprises with more than 10,000 or fewer than 1,000 employees achieved better scores for application performance management best practices than those with between 1,000 and 10,000 employees (see figure below). Why should that be? Here's our theory about why big and small enterprises are doing better job of application performance management than their mid-sized counterparts.
NetForecast's last application performance management (APM) best practices survey showed that those who follow the best application performance management practices experience much better application response times, faster performance problem resolution times, and improved application availability. They are also more likely to uncover application performance problems before users call the help desk, and the performance of their applications is more likely to meet their business needs.
Read more
This week it is Packeteer’s customers’ turn to describe their experiences with the Packetshaper and iShaper distributed application delivery system (DADS) products. This is the last chance for the world to hear about your experiences with Packeteer-branded products, because Blue Coat is set to acquire Packeteer any day now. We know that there are a lot of Packeteer customers out there--more than 10,000 of you according to official sources--and we hope to hear from many of you.
Read more
Do you want to know how well you are managing the performance of your critical applications? Then benchmark! By benchmarking application performance management (APM) best practices you discover how well you are meeting your own performance objectives and how well you stack up against others. A benchmark score shows on a numerical scale how well you are implementing the four best practices we described on May 29 and 30. Benchmarking allows you to compare your own best practice implementations with the industry norm and with those who are executing well and achieving best results.
NetForecast's 2007 APM Benchmarking Survey
Read more
Picture this. You spend a morning looking over the shoulder of an SAP user we'll call Shirley as she enters customer order information, and you watch as around 11 AM Shirley starts to fidget and finally takes a coffee break when her data input task completion times, which you are timing with a stop watch, exceed 10 seconds. After she takes her break, you head back to your cube and over the course of the afternoon and subsequent days you run synthetic tests proving that during peak network usage times SAP task completion times routinely exceed 10 seconds.
Read more
Good application performance management requires structure, and like a well built physical structure, application performance management requires good building blocks that fit together well. The application performance management structure starts with a framework that defines a set of processes that are then delivered using best practices.
Think of the structure in terms of making a great meal. In a culinary context:
Framework = Cuisine
Process = Recipe
Best Practices = Cooking
Read more
For the third installment in our "ADS Customers Speak" series we ask Juniper customers to share their experiences with Juniper's WX solution.
Juniper got into the distributed (a.k.a. dual-ended) ADS market by buying Peribit, the layer 3 compression and caching pioneer. Juniper integrated the Peribit product in 2005 and renamed it the Juniper WX. Juniper's WX application acceleration solution optimizes traffic flows by applying compression and caching techniques, quality of service (QoS) enforcement, bandwidth management, path optimization, and TCP acceleration, as well as acceleration of HTTP, Microsoft CIFS and Microsoft MAPI-based applications.
Read more
We are intrigued by a stealthy little British company called Zeus Technology that recently emerged on the US scene with a software application front end (centralized ADS). Developed by engineers in what those of us who live in Massachusetts call “the other Cambridge”, ZXTM (Zeus Extensible Traffic Manager) runs on standard Intel machines and is easily slipped via VMware into an application’s path to speed it on its way.
Read more
A common theme emerged during several briefings with distributed (a.k.a. dual-ended) ADS vendors at spring Interop--let the voice of the customer be heard. Last Wednesday we posted our first "customer voice" blog, and today we ask Cisco WAAS customers to tell the community about their experiences with the technology.
Read more
We recently caught up with BSM blogger, guru, and evangelist Doug McClure for his thoughts on our blog last week about the need for BSM "Lite". He wholeheartedly agrees with us that the world needs a simpler, less expensive, more responsive way of achieving BSM. Here's what he told us about how to implement BSM Lite.
"Like ITIL, BSM is about process not products. Nearly every client around the world gets enamored by dashboards to bring order to the chaos of IT alignment, but they leave the most important legs of the three-legged stool aside--the people and the process." McClure says that individual IT team members usually focus on narrow world views, and virtually no one in US-based enterprises has end-to-end service ownership.
Read more
Business Service Management (BSM) is the holy grail of good IT management, and achieving it using the current crop of IT management tools from the "big four" management platform vendors is about as hard as finding the real holy grail. The world needs a simpler, less expensive, more responsive way of achieving BSM.
First, let's define BSM. There are several definition sources, the most comprehensive of which is found on Doug McClure's blog. We were surprised to learn that despite the fact that he works for IBM Tivoli, he presents a clear headed and not a product driven BSM view (probably due to his hands-on experience in many BSM engagements).
Read more
While we were at spring Interop we heard a consistent theme from application delivery vendors that went something like this: "Other vendors say things about our products that aren't true to make them look better. Customers love our products and they can tell you the truth." So starting now and continuing every Wednesday for the coming weeks, we want to hear the voices of each major ADS vendor's customers - one vendor at a time. Today we start with Riverbed's Steelhead product--only because Riverbed is the largest pure-play WAN optimization vendor.
Read more
What if everybody called a robin by a different name? It would be downright annoying for birdwatchers to talk about robins because folks wouldn’t know if they were talking about the same bird. They would first have to describe the bird before they could be sure they were referred to the same critter. The same is true with application delivery system (ADS) technologies. Every vendor calls its technology by a different name, so there is no way to compare ADS solutions without first wasting a lot of time describing them.
Read more
Today we dive deeper into the differences between centralized and distributed performance optimization solutions. The technical functions that Centralized and Distributed ADS solutions perform are often very similar—even identical. But there are major differences in the types of traffic that benefit and the span (or reach) of the two solution types. Here are two pictures to help you understand the difference.
Folks often get confused about the difference between centralized (a.k.a., asymmetrical or single ended) and distributed (a.k.a., symmetrical or dual-ended) application performance solutions.
Read more
With so many products claiming to help applications perform well over WANs, how do you know which solutions solve which problems? To bring order to the perplexing chaos of product offerings, we have created a taxonomy for Application Delivery Systems (ADSs). Over the course of the coming weeks in our blog we will fit each of the technologies into this taxonomy and give you a thumbnail sketch of what problem each technology solves and how it works.
Read more
Some would have us believe that the die is cast in the WAN optimization market - that the winners have been chosen, and there are no up and comers. After talking to as many vendors as we could fit into two days at Interop, we beg differ. There are some very interesting strategies and plans in the works that we predict will change the current market dynamics.
Several things are percolating to keep the market interesting - here is an early list:
Read more
Given the uncertain state of the economy, we didn't know what to expect at Interop this spring. The show appears healthy. There is a lot of energy in the crowds, and attendees appear to be paying rapt attention to what is going on at the show. We see lots of attendees filling seats to listen to booth presentations - as if they are here to learn not just enjoy an employer-sponsored junket. For their part exhibitors are more businesslike than in past years, dispensing with gimmicks and diligently trolling for serious leads.
Read more
Last week we posted a simple equation to explain the basic factors influencing application response time. Because it is pared down to show only the most influential factors and not all the factors affecting performance, our simple equation uses an approximation sign. For those of you itching to dive into the bowels of application performance, today we present a much more detailed equation that uses the equal sign. If you really want to understand why your application response times are what they are, read on.
Here's the application response time equation for the number lovers among our readers.
Richard Campbell and Kent Alstad at Strangeloop Networks use the NetForecast response time equation to explain ASP.NET performance issues in an article entitled "Scaling Strategies for ASP.NET Applications" published on the Microsoft MSDN Magazine web site.
The article discusses user application performance satisfaction (response time) and the throughput needed to support many users (scale). They propose ways to measure each key equation element in an operational system and improve its value.
Read more
NetForecast is an internationally recognized engineering consulting company that benchmarks, analyzes, and improves the performance of networked data, voice, and video applications.
|
|
