-
Notifications
You must be signed in to change notification settings - Fork 50
DOC: Supported Virtual Applications
ibmcb edited this page Jan 31, 2013
·
41 revisions
| Type | Topology | Profiles | Data Generation | Load Generator | Load Level | Reported Metrics | Authors |
|---|---|---|---|---|---|---|---|
| DayTrader1 |
1Xclient_daytrader V 1xWAS V 1XDB2 |
single ("default") | Pre-generated | iwlengine | Number of simultaneous clients |
Latency Throughput |
Marcio Silva, Michael Hines |
| Tradelite2 |
1Xclient_tradelite V 1xWAS |
single ("default") | Generated for each new load level | iwlengine | Number of simultaneous clients |
Latency Throughput |
Marcio Silva, Scott Trent |
| LOST3 |
1Xclient_lost V 1xlostfront V 1Xlostback |
single ("default") | Generated for each new load level | autobench | Number of simultaneous clients |
Latency Throughput |
Michael Hines |
| SPECWeb4 |
1Xclient_specweb V 1xspecwebfront V 1Xspecwebback |
single ("default") | Pre-generated | specweb | TBD |
Latency Throughput |
Michael Hines |
| SPECJbb5 | 1Xspecjbb |
single ("default") | Generated for each new load level | specjbb | TBD | Latency and Throughput | Marcio Silva, Michael Dawson |
Comments:
1 (Daytrader)
- DayTrader is a benchmark application built around the paradigm of an online stock trading system. It has 3 tiers, a client VM, generating requests, a WAS VM as the application server and a DB2 VM as the database backend.
- The application performance is reported by the Load Generator process running on the client VM - iwlengine - at the end of an execution with each load level (the execution runs for the duration specified by the load duration).
- Daytrader has two pre-generated database sizes: 400 MB, by setting the attribute "TRADEDB_SIZE = small" and 6 GB, by setting the attribute "TRADEDB_SIZE = large".
- When using the "large" database, the number of users and quotes on the database can be increase by changing the attributes "NR_QUOTES=6000000" (default is 40000) and "NR_USERS=3000000" (default is 15000)
- The database can be placed on a ramdisk inside the DB2 VM, by setting the attribute "DB2_ON_RAMDISK = $True"
- Daytrader can be used in "scale-out" mode, by having a load balancer in front of multiple WAS VMs. To do so, changing the attribute "LOAD_BALANCER = $True". The number of WAS VMs can the be controlled by changing the attribute "SUT=client_daytrader->3_x_was->db2
2 (Tradelite)
- Tradelite uses the same "DayTrader" application and the same Load Generator process (iwlengine) as the above Virtual Application.
- Contrary to the DayTrader Virtual Application, the database is rebuilt each time a new load level is selected, using the default configurations on application.
- Usually Tradelite uses a Derby database, running co-located within the WAS VM. A 3rd tier, containing an actual DB2 database running on a separate VM can be added by setting the attribute "SUT=client_tradelite->was->db2"
3 (LOST)
- LOST stands for (IBM) "LTC Online Store Test". It is basically a 3-tier LAMP (Linux Apache MySQL PHP) stack that simulates and on-line store.
- The whole setup is controlled by autobench to start/stop the Apache server running on lostfront VM and MySQL server running on the lostback VM. It means that whenever a load level is changed by CBTOOL, both Apache and MySQL are restarted. We are looking into change this behavior.
- The application performance is not reported by the Load Generation process. It has to be collected, at regular intervals, from Apache and MySQL.
4 (SPECWeb)
- SPECWeb is a 3-tier application, with a client VM, generating requests, a web server VM and "backend emulator" VM.
- The application performance is not reported by the Load Generation process. It has to be collected, at regular intervals, from Apache running in the specwebfront VM.
- A modified version of the SPECWeb benchmark is used for this Virtual Application, which runs only on 32-bit VMs.
5 (SPECJbb)
- TBA
| Type | Topology | Profiles | Data Generation | Load Generator | Load Level | Reported Metrics | Authors |
|---|---|---|---|---|---|---|---|
| Hadoop6 |
1Xhadoopmaster V 3xhadoopslave |
terasort | Generated once during deployment | terasort | Size of the dataset to be sorted | Latency | Marcio Silva, Yoonseo Choi |
Comments:
6 (Hadoop)
- TBA
| Type | Topology | Profiles | Data Generation | Load Generator | Load Level | Reported Metrics | Authors |
|---|---|---|---|---|---|---|---|
| HPCC7 |
1Xfen_hpc V 3xcn_hpc |
single ("default") | Generated for each new load level | hpcc | N/A |
Latency Throughput Bandwidth |
Marcio Silva, Diego Gallo |
Comments:
7 (HPCC)
- TBA
| Type | Topology | Profiles | Data Generation | Load Generator | Load Level | Reported Metrics | Authors |
|---|---|---|---|---|---|---|---|
| netperf8 |
1Xnetclient V 1xnetserver |
TCP_STREAM TCP_MAERTS TCP_RR TCP_CC TCP_CRR UDP_STREAM UDP_RR |
N/A | netperf | N/A |
Latency Throughput Bandwidth |
Marcio Silva |
| coremark9 | 1Xcoremark |
single ("default") | N/A | coremark | Number of operations to be executed | Latency, Throughput | Marcio Silva |
| filebench10 | 1Xfilebench |
fileserver oltp_noism varmail videoserver webproxy |
Generated for each new load level | filebench | N/A | Latency, Throughput | Marcio Silva |
| ddgen11 | 1Xddgen |
single ("default") | Generated for each new load level | dd | Number of blocks written | Bandwidth | Marcio Silva |
| nullworkload12 | 1Xtinyvm |
single ("default") | N/A | N/A | N/A | N/A | Marcio Silva |
Comments:
8 (netperf)
- TBA
9 (coremark)
- TBA
10 (filebench)
- TBA
11 (ddgen)
- TBA
12 (nullworkload)
- TBA