1 / 11

Justin Dolske, Gojko Babic, Arjan Durresi, Raj Jain

97-0833:Test Experiences and Modifications to Mean Frame Burst Size (MFBS) Section of Performance Testing Baseline Text. Justin Dolske, Gojko Babic, Arjan Durresi, Raj Jain The Ohio State University Contact: Jain@cse.ohio-state.edu http://www.cse.ohio-state.edu/~jain/. Overview. MFBS Review

thom
Download Presentation

Justin Dolske, Gojko Babic, Arjan Durresi, Raj Jain

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. 97-0833:Test Experiences and Modifications to Mean Frame Burst Size (MFBS) Section of Performance Testing Baseline Text. Justin Dolske, Gojko Babic, Arjan Durresi, Raj Jain The Ohio State UniversityContact: Jain@cse.ohio-state.edu http://www.cse.ohio-state.edu/~jain/

  2. Overview • MFBS Review • MFBS Measurement Procedure and Calculation • Measurement Results and Analysis • Summary

  3. MFBS Review • MFBS is the maximum number of frames that each of the source can send at peak rate through switch without incurring any loss. • MFBS measures the data buffering capacity of the switch and its ability to handle back-to-back frames. • The baseline states that MFBS should be measured using k-to-1 configuration. • The experimental results indicate that the MFBS for various values of k can be obtained from the measured MFBS for one k.

  4. MFBS Measurement Procedure and Calculation Source 1 Source 2 Destination Source 3 Source 4 MFBS Test Configuration

  5. MFBS Measurement Procedure and Calculation (contd.) • First, measure MCBS, which is the largest number of back-to-back cells that the sources may send simultaneously without losses. • In order to measure MCBS: • Simultaneously generate fixed length bursts of back-to-back cells through all k VCCs. • Vary the burst length until cell loss is verified at least in one VCC. • Then based on MCBS results, calculate MFBS of frames with a given size.

  6. Traffic Configuration MCBS (per source) 2-to-1 9,050 cells 3-to-1 4,650 cells 4-to-1 3,050 cells Measurement Results for MCBS

  7. Config- 64B 1518B 9188B 64kB uration Frames Frames Frames Frames 2-to-1 434,368 434,148 431,836 393,216 3-to-1 223,168 223,146 220,512 196,608 4-to-1 146,368 145,728 137,820 131,072 MFBS [in Bytes] Calculated from Measured MCBS

  8. Analysis of Measurement Results • Given (measured) MCBS for k, MCBSfor j can be calculated as: MCBSj = MCBSk *(k-1)/(j-1) Use k = 2 to minimize the equipment needed. Configuration The Switch in each interval transmits: buffers: 2-to-1 1 1 3-to-1 1 2 4-to-1 1 3 i-to-1 1 i-1

  9. Traffic Characteristics • Type of VCCs: permanent virtual path connections, switched virtual path connections, permanent virtual channel connections, switch virtual channel connections; • VCCs established: between ports inside a network module, between ports on different network modules, between ports on different fabrics, some combination of previous cases; • Connection configuration: 2-to-1; • Frame length: 64 B, 1518 B, 9188 B, 64 kB;

  10. Summary • A procedure to obtain the MFBS is defined. • 2-to-1 configuration is sufficient • Detailed traffic characteristics are defined.

  11. Motion • Adopt the text of 97-0833 to replace section 3.5 of Performance Testing Baseline Text.

More Related