ARA Networks, PolyMix-4

The Measurement Factory

In January 2002, The Measurement Factory benchmarked C3300, a caching proxy offered by ARA Networks. The tests were executed in our Boulder lab using the PolyMix-4 workload.

This report also shows the results for a cache-off entry running ARA Networks's Jaguar software, labeled Aratech C3100 in the cache-off report. Thus, this report covers all PolyMix-4 results for ARA Networks offerings available to date.

We have followed the fourth cache-off rules and methodology to ease comparison of the results. This short report does not describe the workload or the caveats of interpreting the results. Familiarity with the official cache-off report is a must!

We also note that while the workload for this test is identical to what we used during the fourth cache-off, it is somewhat unfair to compare competitors results at the cache-off with the results achieved in a lab: Cache-off participants do not know each-other results prior to the event and also may have improved the performance of their products since the cache-off.

As always, we do not give purchasing advise. We provide performance measurements and expect the reader to use this information, along with other important factors, in their decision making process.

Table of Contents

1. Executive Summary
2. Where to find more information
3. Product Configurations
4. Vendor Comments

1. Executive Summary

The ``Executive Summary'' table below summarizes the performance results.

Baseline PolyMix-4 Results
Product Total
Price

(US$)
Peak
Throughput
Response Time
(sec)
Savings
(%)
$1,000
can buy
Minimum
Downtime

(minutes)
Cache
Age

(hour)
req/sec Mb/sec Hit All Miss Xact Bwidth hit/sec req/sec
C3100 14900 1500 83 0.101 1.40 2.70 51.18 33.60 52 101 3.07 17.45
C3300 22400 2800 152 0.025 1.28 2.66 52.53 37.96 66 125 n/a 29.68

Links in column headers point to bar charts comparing the corresponding measurement. Product links point to a auto-generated individual pages for the entries.

All performance tests finished with negligible number of errors.

Downtime test result is not available for C3300, though we observed the cache to take at least 3-5 minutes to become operational, due to large number of SCSI disks that are spinned and probed sequentially.

The C3300 test was executed on ARA Networks's request. The same entry we tested in January was entered in the cache-off competition in November, but was not ready to be benchmarked at that time. According to the cache-off rules, The Measurement Factory offered ARA Networks to re-test the entry shortly after the competition. The declared price and configuration of the C3300 entry has not changed since November.

2. Where to find more information

The Measurement Factory maintains the Official Results Site, where this report and the detailed Polygraph log files from all tests are stored. All information at the Official Results Site is freely available.

Documentation, sources, discussion mailing lists, and other information related to Polygraph benchmark are available at the Web Polygraph site.

We discuss only major performance measurements in this report. If you'd like more details about a particular entry, consult these resources:

3. Product Configurations

Here are the configuration details for all tested products. Note the change in software: cache-off entry was running Jaguar2000 and Jaguar3000 was used in January.

Label Full product name Price
(US$)
Avail
able
(mm/yy)
CPU
(MHz)
RAM
(MB)
Cache disks
(n · GB)
Rack
Space
(RU)
Cache
(GB)
Software
C3100 ARA Networks C3100 14900 01/02 1000 1024 5 · 36 3 180 Jaguar2000
C3300 ARA Networks C3300 22400 02/02 1200 3072 10 · 36 5 308 Jaguar3000

4. Vendor Comments

It is our tradition to give vendors a chance to comment on the results after they have seen the review draft. The comments below are verbatim vendor submission. The Measurement Factory has not verified any of the claims, promises, or speculations that these comments may contain.

AraNetwork Technologies
http://www.aranetwork.com

ARA Networks participated in the 3rd Cache-off with JAGUAR2000. Since then, JAGUAR2000 has been consistently redeveloped for better performance and versatile functions. In this test, ARA Networks introduced JAGUAR3000, ARA Networks next generation cache server, and it demonstrated the worlds leading performance. JAGUAR3000 distinguished itself with its excellent performance and diverse functions meeting various market needs.

JAGUAR servers proved to have the best peak throughput in both 100M and 1000M peers with a high hit ratio and low response time. Armed with MCT (Minimal Context-switching Thread, ARA Networks' original architecture), JAGUAR3000 runs in kernel mode, utilizes multiple CPUs and breaks the performance barrier of today's cache engines. API(SDK) will also be provided to easily integrate the various valued-added solutions of 3rd parties with JAGUAR3000.

ARA Networks, an ITM company, develops and provides streaming CDN, dynamic caching, browser caching and web caching solutions. ARA Networks provides beneficial partnership programs and a flexible price policy.

We'd more than appreciate your continued interest in ARA Networks, which advances every year through the annual Cache-off.