Month: February 2006

GrabPERF: State of the System, Feb 2006

If anyone wants to know why I am proud of GrabPERF, this graph should give you a clue.

GrabPERF Growth

Every hour, an aggregated value is produced for every test url. Counting up the Geometric Mean aggregations on a daily basis, the growth line is pretty amazing.

Doing a rough calculation, the system has grown from testing 40 urls to testing 104 urls.

But the true scope of this growth can only be seen by looking at the number of data insertions into the raw data table on a daily basis.

GrabPERF Measurements Per Day

Currently, with the four measurement agents (I turned down one of the Technorati agents today to relieve the strain on the database server), the system is handling nearly 300,000 data insertions a day. Not even in the same timezone with most of the major sites I measure, but when I think that this is a system that I designed, I am stunned. For an unfunded, not-for-profit, one-person effort, I am constant astounded by what this system can handle.

Other areas of note over the last year:

  • The Technorati-donated servers can now easily hold 60 days of detail data on a robust enterprise grade system, in a real datacenter
  • The basement datacenter is now closed
  • The new interface was created specifically to allow the system to grow and easily accommodate new features
  • People are now approaching me on a daily basis to have sites added, or to have data explained
  • GrabPERF went from one measurement agent to five agents in four locations, including a location in Europe (Portugal)
  • GrabPERF has been used in various places to serve as an indicator or motivator for performance improvement, including: Bloglines, Ping-O-Matic, Technorati, PubSub (1 and 2), Blogwise, and others.
  • A number of corporate and individual sponsors have stepped forward to support our efforts in many different ways: money, servers, measurement locations, commentary and critiques

I know that I have said it many times in person, or online, but thank you all. Those of you who use and support GrabPERF are the ones who continue to motivate me to make this system better on a daily basis.

Keep those cards and letter coming.

Technorati Tags: , , ,

DNS | Apache Virtual Host Madness Today

I have noticed that GrabPERF has been responding increasingly more slowly as of late. Well, I believe that I have resolved the performance issue: I moved the Web component of GrabPERF off of the machine where the database is housed.

However, when I did this, I hit a really stupid issue that was the result of a legacy httpd.conf file directive.

As well as GrabPERF, I moved this blog, also a heavy HTTP | PHP user, onto the same Web server. Then, once I had seen the DNS propagate, I went to this blog….and got the GrabPERF homepage!

WTF!?!?!?!

Turns out that I was the victim of a REALLY dense mis-configuration, which I removed from the new Web server configuration file. I had buried the NameVirtualHosts directive in a VirtualHost container, which was not part of the new server’s config file.

Without the NameVirtualHosts directive set, the server happily responds to all incoming requests with the first VirtualHost it finds in the httpd.conf file, which in in this case was GrabPERF.

Once I solved this, and placed the NameVirtualHosts directive outside of all of the VirtualHost containers, the server began working perfectly.

I then went and retrofitted the secondary Web server.

If none of this makes sense, it’s ok. I am not feeling real lucid right now.

Technorati Tags: , , ,

Opera Web Site: Hello? Anyone home?

As a Web performance analyst, this is the sort of graph that melts your mind and makes you want to scream at the gods|goddesses to rain a plague of locusts on the offending System|Network|Web Site Administrators.

Opera Server Errors - Feb 17 2006
Click for a larger version

This is a classic display of the “run in circles, scream and shout” problem identification and resolution process.

One-half of the Opera Web servers at http://www.opera.com/ are not responding.

50%.

They have three separate public IP addresses, and it is clear that behind those IPs are many more NAT|VIP machines that respond to requests.

Did I mention that one-half of the machines at all three IP addresses are not responding to HTTP requests?

Has anyone at Opera noticed this?

Sometimes I feel like even the echoes aren’t listening anymore.

Technorati Tags: , , , ,

AJAX and Web Performance Improvement: How do you measure it?

AJAX is an amazing bit of technology, and a boon to Web performance.
The question is, how do you accurately measure it?
Now, from the perspective of synthetic transaction measurement, AJAX destroys the foundational concept of the “Page”, as it is built on the concept of the “sequence”. “Pages” assume a whole new HTML document is loaded in each step, where a “sequence” simply tracks the specific flow of steps that the customer performs, regardless of whether they occur in a new HTML document or not.
In this respect, passive monitoring services currently have a distinct advantage over synthetic measurement, because they intrinsically track the sequence of event that a customer triggers, rather than the pages that are downloaded. I will not declare the death of synthetic Web performance measurement yet (my day job is with one the largest synthetic Web performance measurement companies); but the industry has to re-evaluate many of its core precepts.
AJAX is a technology that will definitely benefit from the blending of passive and synthetic performance measurement into a single usable stream of information that companies can act on. With this blend, companies can determine how their servers are responding, as well as what customers are doing, tracking the flow of business data in real-time.
What will be interesting to watch is how the synthetic measurement companies (including the one I work for) respond to this. One of the companies in our space says that they handle it now, but I have yet to see the results of their effort and how they really implement it in the field.
How are you and your team measuring the performance of your AJAX applications?
Technorati Tags: , , , ,

AJAX has a positive effect on Web Performance

Or so it seems from this article!

Makes sense if you think of it. Only part of the page is updated, so less bandwidth is used. And if you compress that data as well, you save even more.

Maybe AJAX is not just a novelty.


UPDATE [Feb 17 2006]: Some great additional articles are showing up in the comments. The catalogue demonstration is very cool.


UPDATE [April 29 2022]: AJAX, in the form of Single Page Application (SPA) Frameworks won.

Copyright © 2024 Performance Zen

Theme by Anders NorenUp ↑