Changes between Version 25 and Version 26 of Performance
- Timestamp:
- 2008-01-30T19:04:18Z (17 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Performance
v25 v26 62 62 [/tahoe-figleaf-graph/hanford.allmydata.com-tahoe_speedstats_rate.html rate graph] show these Ax+B numbers for a node in colo and a node behind a DSL line. 63 63 64 The [/tahoe-figleaf-graph/hanford.allmydata.com-tahoe_speedstats_delay_SSK.html mutable-file delay graph] shows the "B" per-file latency number65 for mutable (aka "SSK") files. In the 0.7.0 release, this is dominated by the RSA keypair generation necessary to create each new mutable file.66 67 64 The 68 65 [/tahoe-figleaf-graph/hanford.allmydata.com-tahoe_speedstats_delay_rtt.html delay*RTT graph] shows this per-file delay as a multiple of the average round-trip … … 71 68 per-file delay in units of RTT helps to compare the observed performance 72 69 against the predicted value. 70 71 === Mutable Files === 72 73 Tahoe's mutable files (sometimes known as "SSK" files) are encoded 74 differently than the immutable ones (aka "CHK" files). Creating these mutable 75 file slots currently (in release 0.7.0) requires an RSA keypair generation. 76 [http://allmydata.org/tahoe-figleaf-graph/hanford.allmydata.com-tahoe_speedstats_SSK_creation.html This graph] 77 tracks the amount of time it takes to perform 78 this step. 79 80 There is also per-file overhead for upload and download, just like with CHK 81 files, mostly involving the queries to find out which servers are holding 82 which versions of the file. The 83 [/tahoe-figleaf-graph/hanford.allmydata.com-tahoe_speedstats_delay_SSK.html mutable-file delay graph] 84 shows this "B" per-file latency value. 85 86 The "A" transfer rate for SSK files is also tracked. TODO: add a link to the 87 graph. 73 88 74 89 === Roundtrips ===