37
7
4
x 10
x 10
2
5
DISK
Local Cache
1.5
4
Remote Cache
DISK
Local Cache
3
1
Remote Cache
2
Throughput
0.5
Accumulated Latency (ms)
1
0
0
10
9
8
7
6
5
4
3
10
9
8
7
6
5
4
3
k
k
(a) Accumulated latency
(b) Throughput
Fig. 3.7.
Breakdown of Accumulated Latency and Throughput of a 32 node Web
Cluster. Three Bars Represent the Results of the press via, dcs, adaptive Models from
Left to Right
of each type of requests, local cache, remote cache and disk, when k is 10 and 5. At
k = 10, it shows that the average latency of remote cache requests in the press via
model is about four times longer compared to that of either the dcs or adaptive model.
The local cache latency of dcs is slightly longer compared to the press via because
the remote cache read requests are served with higher priority than local cache read
requests in the dcs model.
However, because the adaptive model employs the same coscheduling scheme as
dcs, the average latency of the local cache read requests are only slightly shorter. This
is because when the adaptive model has a large number of local cache read requests to
process, the communicating processes cannot raise an interrupt to preempt the CPU.
In addition, since it can process the remote cache read requests more efficiently than
other requests, the CPU has more time to handle other types of requests. At k = 5, as












  

Home

About Services Network Support FAQ Order Contact
 

 

Clan Web Hosting

Our partners:Jsp Web Hosting Unlimited Web Hosting Cheapest Web Hosting  Java Web Hosting Web Templates Best Web Templates PHP Mysql Web Hosting Interland Web Hosting Cheap Web Hosting PHP Web Hosting Tomcat Web Hosting Quality Web Hosting Best Web Hosting  Mac Web Hosting 

Lunarwebhost.net  Business web hosting division of Vision Web Hosting Inc. All rights reserved