Changes between Version 40 and Version 41 of FGBI


Ignore:
Timestamp:
10/07/11 00:08:27 (13 years ago)
Author:
lvpeng
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • FGBI

    v40 v41  
    5151Figures 2a, 2b, 2c, and 2d show the type I downtime com-
    5252parison among [wiki:FGBI FGBI], [wiki:LLM LLM], and [http://nss.cs.ubc.ca/remus/ Remus] mechanisms under [http://httpd.apache.org/ Apache], [http://www.nas.nasa.gov/Resources/Software/npb.html NPB-EP],
    53 SPECweb, and SPECsys applications, respectively. The block size used in all
     53[http://www.spec.org/web2005/ SPECweb], and SPECsys applications, respectively. The block size used in all
    5454experiments is 64 bytes. For [http://nss.cs.ubc.ca/remus/ Remus] and [wiki:FGBI FGBI], the checkpointing period is the
    5555time interval of system update migration, whereas for [wiki:LLM LLM], the checkpointing
     
    6565worse than [http://nss.cs.ubc.ca/remus/ Remus]).
    6666
    67 Although SPECweb is a web workload, it still has a high page modifi-
     67Although [http://www.spec.org/web2005/ SPECweb] is a web workload, it still has a high page modifi-
    6868cation rate, which is approximately 12,000 pages/second. In our experi-
    6969ment, the 1 Gbps migration link is capable of transferring approximately 25,000
    70 pages/second. Thus, SPECweb is not a lightweight computational workload for
     70pages/second. Thus, [http://www.spec.org/web2005/ SPECweb] is not a lightweight computational workload for
    7171these migration mechanisms. As a result, the relationship between [wiki:FGBI FGBI] and
    7272[wiki:LLM LLM] in Figure 2c is more similar to that in Figure 2b (and also Figure 2d),
     
    7474downtime by as much as 77%. Moreover, compared with [http://nss.cs.ubc.ca/remus/ Remus], [wiki:FGBI FGBI] yields a
    7575shorter downtime, by as much as 31% under [http://httpd.apache.org/ Apache], 45% under [http://www.nas.nasa.gov/Resources/Software/npb.html NPB-EP], 39%
    76 under SPECweb, and 35% under SPECsys.
     76under [http://www.spec.org/web2005/ SPECweb], and 35% under SPECsys.
    7777
    7878[[Image(table1.jpg)]]
     
    100100
    101101Figure 3a shows the overhead during VM migration. The figure compares the
    102 applications' runtime with and without migration, under [http://httpd.apache.org/ Apache], SPECweb,
     102applications' runtime with and without migration, under [http://httpd.apache.org/ Apache], [http://www.spec.org/web2005/ SPECweb],
    103103[http://www.nas.nasa.gov/Resources/Software/npb.html NPB-EP], and SPECsys, with the size of the fine-grained blocks varies from 64
    104104bytes to 128 bytes and 256 bytes. We observe that in all cases the overhead is