1 / 14

Evaluation of Objectivity/AMS on the Wide Area Network

Evaluation of Objectivity/AMS on the Wide Area Network. SATO Hiroyuki KEK Computing Research Center. Introduction. Measured the performances of Objectivity/AMS : AMS write/read access (5.1 vs 5.2) AMS over LAN and WAN

turi
Download Presentation

Evaluation of Objectivity/AMS on the Wide Area Network

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Evaluation of Objectivity/AMSon the Wide Area Network SATO Hiroyuki KEK Computing Research Center

  2. Introduction • Measured the performances of Objectivity/AMS : • AMS write/read access (5.1 vs 5.2) • AMS over LAN and WAN • Resource utilization of the AMS server is monitored for 5.1 and 5.2 on LAN • CPU utilization • TCP/IP packet was monitored for AMS transaction over LAN and WAN. • Transfer rate

  3. Testbed Configuration Monitored CPU/Packet/…

  4. AMS/Write-Read Performance Test on LAN and WAN • Objectivity page size is 8192 bytes • Test object (40 bytes) : 175~176 objects / page • #’objects for the write/read : • 1,000,000 objects for LAN : 570 pages • 100,000 objects for WAN : 57 pages

  5. 40 40 30 30 CPU Utilization (%) CPU Utilization (%) Write 20 20 10 10 0 0 0 0 20 20 40 40 60 60 Time (sec) Time (sec) 30 30 20 20 CPU Utilization (%) CPU Utilization (%) Read 10 10 0 0 0 0 20 20 40 40 60 60 80 80 100 100 Time (sec) Time (sec) V5.1 V5.2 CPU Utilization for Write/Readfor V5.1 and V5.2 on LAN AVE:12% AVE:18% AVE:4% AVE:8%

  6. Round Trip Time between KEK and CERN 500 (msec) 400 300 0 10 20 30 Day in 2000 Measurements on the WAN • Characteristics of the network between KEK and CERN • RTT (Round Trip Time) is about 300 msec • The bandwidth is 2Mbps

  7. Server (AMS) at KEK Client at KEK/CERN 1200 1200 1000 1000 52 bytes 800 800 Transfer Rate (kB/s) Transfer Rate (kB/s) 600 600 LAN 8236 bytes 400 400 8280 bytes 200 200 0 0 36 bytes Time (sec) 8280 bytes WAN 0 0 20 20 40 40 60 60 80 80 (sec) Time Data Transfer for Write on LAN and WAN Control Transfer Phase (CTP) Data Transfer Phase (DTP) Bulk Transfer

  8. Server (AMS) at KEK Client at KEK/CERN 500 500 52 bytes 400 400 Transfer Rate (kB/s) Transfer Rate (kB/s) RTT 8236 bytes LAN 300 300 200 200 100 100 0 0 0 25 50 75 100 Time (sec) x 103 4200 Write Seq. Number 4100 WAN Read 4000 0 1 2 0 50 100 150 200 Time (sec) Time since #’4M is sent (sec) Data Transfer for Readon LAN and WAN Control Data Transfer Phase (CDTP)

  9. 140 120 100 80 60 40 20 0 Comparison with AMS Writeand ftp Write performance over WAN is faster than ftp ! Transfer Rate (kB/s) AMS ftp 0 20 40 60 80 (sec) Time

  10. Window Size in TCP/IP • Monitored with tcpdump • AMS : 33580 bytes • ftp : 24820 bytes • Window size affects the transfer rate • It can be changed with “setsockopt” function (standard max : 65535 bytes / optional 232 bytes) • Bulk data transfer rate for various received window size is measured

  11. 02/Feb/2000 250 Window size 200 64240 bytes (Test Program) Data Transfer Rate (kB/s) 150 33580 bytes (Objectivity) 100 50 24820 bytes (ftp) 0 0 6 12 18 24 (hour) Time Bulk Data Transfer Rate for Various Window Size (KEK - CERN)

  12. 10 2 10 1 Window Size RTT Bandwidth 1 Efficiency = 10 -1 10 -2 10 3 10 4 10 5 10 6 10 7 Effective Bandwidthfor RTT = 280 ms (KEK - CERN) ftp AMS Efficiency (%) 2Mbps 20Mbps Needs the window scale option 200Mbps Window Size (bytes)

  13. CERN 1 KEK Read 1 2 2 3 3 4 satellite (RTT=655ms) 4 5 3 5 6 6 7 7 8 8 9 ack2 2 Transaction Time / Page (sec) 10 10 11 11 ack4 12 TCP Timeout 12 13 1 13 ack6 surface (RTT=300ms) 14 14 15 15 16 16 9 0 9 ack8 100 150 200 ack8 ack8 Page Number ack8 ack8 8236 = 536*15+196 ==> 16 segmets ack8 ack8 MSS for satellite ack16 Comparison with network of surface and satellite LOST! ack8 The congestion window is initialized after this transaction.

  14. Summary • Write/Read operation of Objectivity 5.2 works on the WAN • AMS server 5.2 is multi-threaded • Consumes more CPU cycles than 5.1 • Does not matter when network is a bottleneck • Write performance over WAN is faster than ftp • more speedup is expected with a larger window size (Optimal window size may vary with the available buffer size of the network router) • Read performance is poor due to the hand-shaking • Oscillation of congestion window is observed in satellite network — further study needed

More Related