1. Evaluating Energy and Performance for Server- Class Hardware Configurations Chenguang Liu, Jianzhong Huang, Qiang Cao, Shenggang Wan, Changsheng Xie School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China

2. chgliu2009@gmail.com Challenge Problems Data Centers 15-40x the energy intensity of typical office buildings A single rack of servers can be 20 kW – $17k per year (at $.10/kWh) per rack – Hundreds of racks per center For high-end data centers, energy cost increased as much as 25% annually, and making it a big consideration in the total cost of ownership.

3. chgliu2009@gmail.com Problem Example Strategies (software & hardware) —migrate workloads & shut down —become more energy proportional All of them need qualitative and quantitative reference

4. chgliu2009@gmail.com Add apples or hamburgers? How much to add?

5. chgliu2009@gmail.com Energy & Performance Metrics Energy —the average power(Watts) Performance —I/O operations per second(IOPS) Integrated factor —I/O operations per joule

6. chgliu2009@gmail.com Energy & Performance Info Gather Energy —ZH101 Performance —FileBench

7. chgliu2009@gmail.com Workload Categories cont. FileServer —a file system workload similar to SPECsfs Varmail —a /var/mail NFS mail server emulation WebServer —a mix of open/read/close ops of multiple files in a directory tree OLTP —a database emulator using I/O model from Oracle 9i

8. chgliu2009@gmail.com Workload Properties Workloadfilemailwebdatabase Average file size 64MB16KB64KB10MB Average dir depth 5001501 Number of files 100001000 10 I/O sizes R/W 1MB 2KB Number of threads 5016100200 R/W Ratio1:21:110:11:1

9. chgliu2009@gmail.com Test Bed SpecificationMachine Information Server modelDell R710 Machine age1 yr CPU modelIntel Xeon 5500 CPU speed2.26GHz No. of CPUs2 dual core(adjustable) L1 cache size16K L2 cache size2M L3 cache size8M RAM size4*4G DDR3 (adjustable) RAM typeDIMM Disk TypeSAS Disk Size137G Disk RPM7200 Disk cache8M ChipsetIntel 5520

10. chgliu2009@gmail.com Elements of Total Energy Consumption —W overhead =energy for maintaining the server's fundamental operating mode —W CPU, W mem & W disk =energy for workloads on different hardware parts respectively —α, β & γ=the weight factor

11. chgliu2009@gmail.com Average Power in 6 configurations

12. chgliu2009@gmail.com When a server does not perform any work, it consumes the most energy. The workload selection alone cannot reduce idle power, but combined with right-sizing techniques, it can improve power efficiency by prolonging idle periods. Different workloads exercise the system’s resources differently, directly affecting the additional power

13. chgliu2009@gmail.com Results in FileServer Workload

14. chgliu2009@gmail.com Performance of fileserver workload is both CPU and memory-critical The configuration of 1u8g proved to be the most power-efficient one, achieving the top point of 26.7373(iops/J) Energy of fileserver workload is memory- critical Results in Fileserver Workload (cont.)

15. chgliu2009@gmail.com Results in Varmail Workload

16. chgliu2009@gmail.com Performance of Varmail workload is CPU- critical The configuration of 2u4g proved to be the most power-efficient one, achieving the top point of 15.0094(iops/J) Energy of Varmail workload is also CPU-critical Results in Varmail Workload (cont.)

17. chgliu2009@gmail.com Results in WebServer Workload

18. chgliu2009@gmail.com Performance of webserver workload is neither CPU nor memory-critical The configuration of 1u4g(the least power consumption) proved to be the most power- efficient one Web server’s demand for hardware is easy to meet Results in Webserver Workload (cont.)

19. chgliu2009@gmail.com Results in OLTP Workload

20. chgliu2009@gmail.com Results in OLTP Workload (cont.) Performance of database server workload is neither CPU nor memory-critical Like web server, 1u4g will be a good choice

21. chgliu2009@gmail.com In fact and intuitively, as to the characteristic of OLTP processing, α, β and γ should be great values A reasonable explanation may come from the inner mechanism(database lock) of OLTP applications blocking the contributions of hardware enhancement.

22. chgliu2009@gmail.com Conclusion Strengthening every aspect of a server configuration is not always a wise thing(not just the matter of hardware configuration) Since the computing and storage requirements of web server applications are easy to meet and we can't easily improve the performance of OLTP applications, we can aggregate different workloads on the single server according to the actual situation

23. chgliu2009@gmail.com Conclusion (cont.) Even in single machine different workloads need different hardware configurations to achieve performance- and energy- efficiency In the management of a modern datacenter, a hardware configuration adaptor would play a very significant role

24. chgliu2009@gmail.com THANK YOU Questions?