close
Warning:
Can't synchronize with repository "(default)" (Unsupported version control system "svn": /usr/lib/python2.7/dist-packages/libsvn/_fs.so: failed to map segment from shared object: Cannot allocate memory). Look in the Trac log for more information.
- Timestamp:
-
Sep 29, 2010, 2:56:13 PM (14 years ago)
- Author:
-
jazz
- Comment:
-
--
Legend:
- Unmodified
- Added
- Removed
- Modified
-
v53
|
v54
|
|
63 | 63 | * 實做三 : [wiki:waue/2010/HbaseThrift Hbase + Thrift + PHP ] |
64 | 64 | * 實做四 : [wiki:NCHCCloudCourse100929_4_HBEX HBase程式設計] , [wiki:NCHCCloudCourse100929_4_HEXM 範例] |
| 65 | |
| 66 | = 補充資料 = |
| 67 | |
| 68 | * [benchmark] [http://wiki.github.com/brianfrankcooper/YCSB/ Yahoo! Cloud Serving Benchmark (YCSB)] - [http://research.yahoo.com/node/3202 論文: Benchmarking Cloud Serving Systems with YCSB] |
| 69 | * Yahoo 寫的效能測試軟體,YCSB 1.2 版支援 HBase, Cassandra, MongoDB |
| 70 | * 從[http://www.brianfrankcooper.net/pubs/ycsb-socc-talk.ppt 論文簡報投影片]得到的[http://www.brianfrankcooper.net/pubs/ycsb-v4.pdf 結論]注意到幾件事情: |
| 71 | * 若資料庫特性是一半讀一半寫,MySQL 大約在 7500 operations/sec 時會到達效能極限。若資料庫特性是 95% 讀 5% 寫,就沒看到明顯效能極限。 |
| 72 | * HBase 寫入飛快,但讀取會稍微慢一點,而且讀取的效率不太線性。(寫入記憶體所以超快,但成本也比較高,查詢時重建紀錄) 加機器時,不會有明顯的效能差異 (跟 HDFS 一樣,要跑 balancer 才會搬資料)。 - Low Update Latency, Higher Read Latency |
| 73 | * Cassandra 寫入稍微慢一點,讀取則比 HBase 快,特別是資料庫特性是 95% 讀取時比 HBase 快。加機器的時候,會需要一段時間才能讓效能穩定下來(這部分我覺得是副本架構造成的)。 |
| 74 | * [[Image(NCHCCloudCourse100928:cassandra_hbase_update_heavy.png,width=800)]] |
| 75 | * [[Image(NCHCCloudCourse100928:cassandra_hbase_read_heavy.png,width=800)]] |
| 76 | * [[Image(NCHCCloudCourse100928:cassandra_hbase_short_scans.png,width=800)]] |
| 77 | * [[Image(NCHCCloudCourse100928:cassandra_hbase_scale_up.png,width=800)]] |