This document contains an ongoing benchmark of Hypertable, comparing the performance of the system run on top of KFS and HDFS.

Small Insert Test

Machine Profile

• 1 1.8GHz Dual-core Opteron Processor 2210
• 4 GB RAM
• 4 7200 RPM SATA drives (mounted JBOD)

The Data (AOL Query log)

Here's an example of what the AOL query log data looks like:

AnonID  Query   ItemRank        ClickURL
1636218 www.airtime500.com      2       http://www.airtime500.com
2272416 theunorthodoxjew.blogspot.com   1       http://theunorthodoxjew.blogspot.com
172627  www.yahoolagins.com
2569723 www.homesforsale
1196769 zip codes       1       http://www.usps.com
724416  propertytaxsales.com
30011   schwab learning 1       http://www.schwablearning.org
[...]

The row key used was the AnonID. The Query, ItemRank, and ClickURL, columns were inserted into a table that was created with the following HQL command:

CREATE TABLE "query-log" (
Query,
ItemRank,
ClickURL
);

The query log was sorted by timestamp. Each line of the query log could generate from 1 to 3 inserts, depending on how many column values were present. After the entire log was inserted, the table contained 75,274,825 cells. On average, the size of each row key was about 7 bytes and the size of each value inserted was 15 bytes.

Results

NOTE: HDFS currently does not support an flush (fsync) operation, so to be fair we disabled the flush call for the KFS run.

Eight machines (motherlode001-motherlode008) were used to run both HDFS (version 0.16.4) and Hypertable. The Hypertable master and Hyperspace (Chubby) were run on motherlode001. HDFS was configured to use only three of the four available drives on each machine and was configured with 3-way replication. The log was sorted by timestamp and split into 5 pieces (a-no-timestamps.tsv, b-no-timestamps.tsv, c-no-timestamps.tsv, d-no-timestamps.tsv, e-no-timestamps.tsv. Two separate machines (motherlode000 and motherlode009) were used to run insert clients. The following HQL commands were started more-or-less simultaneously:

motherlode000:
LOAD DATA INFILE ROW_KEY_COLUMN=AnonID 'a-no-timestamps.tsv' INTO TABLE "query-log";
LOAD DATA INFILE ROW_KEY_COLUMN=AnonID 'b-no-timestamps.tsv' INTO TABLE "query-log";

motherlode009:
LOAD DATA INFILE ROW_KEY_COLUMN=AnonID 'c-no-timestamps.tsv' INTO TABLE "query-log";
LOAD DATA INFILE ROW_KEY_COLUMN=AnonID 'd-no-timestamps.tsv' INTO TABLE "query-log";

HDFS (no flush)

 Elapsed time:  170.66 s
Avg value size:  15.25 bytes
 Avg key size:  7.10 bytes
   Throughput:  1792158.60 bytes/s
 Total inserts:  14825279
   Throughput:  86869.79 inserts/s

 Elapsed time:  167.44 s
Avg value size:  15.26 bytes
 Avg key size:  7.11 bytes
   Throughput:  1871062.70 bytes/s
 Total inserts:  15185349
   Throughput:  90690.84 inserts/s

 Elapsed time:  179.91 s
Avg value size:  15.20 bytes
 Avg key size:  7.03 bytes
   Throughput:  1737888.10 bytes/s
 Total inserts:  15208310
   Throughput:  84532.68 inserts/s

 Elapsed time:  169.57 s
Avg value size:  15.22 bytes
 Avg key size:  7.11 bytes
   Throughput:  1831688.52 bytes/s
 Total inserts:  15080926
   Throughput:  88937.45 inserts/s

KFS (no flush)

 Elapsed time:  125.51 s
Avg value size:  15.25 bytes
 Avg key size:  7.10 bytes
   Throughput:  2436864.83 bytes/s
 Total inserts:  14825279
   Throughput:  118120.09 inserts/s

 Elapsed time:  126.25 s
Avg value size:  15.26 bytes
 Avg key size:  7.11 bytes
   Throughput:  2481447.59 bytes/s
 Total inserts:  15185349
   Throughput:  120276.33 inserts/s

 Elapsed time:  135.51 s
Avg value size:  15.20 bytes
 Avg key size:  7.03 bytes
   Throughput:  2307335.26 bytes/s
 Total inserts:  15208310
   Throughput:  112231.19 inserts/s

 Elapsed time:  127.66 s
Avg value size:  15.22 bytes
 Avg key size:  7.11 bytes
   Throughput:  2433069.68 bytes/s
 Total inserts:  15080926
   Throughput:  118137.45 inserts/s

With flush enabled, the KFS insert rate dropped to approximately 85K inserts/s.