ASP.NET Micro Caching: Benefits of a One-Second Cache: ASP Alliance

Print Add To Favorites Email To Friend Rate This Article
ASP.NET Micro Caching: Benefits of a One-Second Cache

page 4 of 7

by Steven Smith
Feedback

Average Rating:

Views (Total / Last 10 Days): 41918/ 261

Article Contents:

Results Without Caching

Running the test without any caching on the web server resulted in 16,961 requests and an average requests per second of 56.54. The Time to First Byte (TTFB) and Time to Last Byte (TTLB), which are useful measures of the site's performance from an individual user's perspective, were both 52ms--instant as far as a user is concerned. For most web applications, anything under one second is an acceptable TTLB.

Below is a graph of requests per second over time for the page without caching.

Figure 1

Performance Without Caching

« (Page 3)

View Entire Article

(Page 5) »

Article Feedback

Comment:
Title:
Name:
Url:	( Optional )

Please add 7 and 3 and type the answer here:

User Comments

Title: Simultaneous requests
Name: The Chief
Date: 5/18/2006 8:10:20 AM
Comment:
Hi

If it takes 0.5 seconds to load data and render the page and the server receives 20 requests within this time, does asp.net hold off processing the other 19 requests for the same page until it has finished rendering and caching the output from the first of these 20 page requests?

Your results would suggest that this is indeed what is happening but would appreciate confirmation of this.

We use caching on a busy e-commerce site, but there is still a bit of a hit on the server when the site restarts as all the output on each different category page is loaded and cached at the same time. I'm wondering whether the same data is being loaded twice via multiple requests.

Thanks

Title: Caching
Name: Arpi
Date: 2/9/2006 6:57:03 AM
Comment:
This article explain right.Hence I want to know how can we stop caching of the system

Title: Not So Silly
Name: Steven Smith
Date: 7/28/2005 8:32:35 AM
Comment:
Not so - consider a news site like CNN.com. With their traffic I'm sure certain articles are read more than 1/second, and I'm willing to bet their articles are stored in a database. The same is true for an eBay auction nearing its end, or a popular Amazon.com book. There are plenty of large scale applications that could theoretically benefit from this technique. That said, you have to test your own application using realistic usage data for your app, then set the cache duration to something that makes sense for you.

Title: Silly
Name: Cookie Monster
Date: 7/28/2005 3:05:54 AM
Comment:
This is silly. Only in this test would a 1 second cache make this large of a difference. In the real world, users wouldn't be accessing the exact same piece of data so often, so basically every request would end up a cache miss.

Product Spotlight