OSS App Hackathon @ National Information Society Agency

Yesterday, there was a OSS App Hackathon arranged by the NIA (National Information Society Agency) in Seoul. I attended as a panel of judges w/ Prof. Lee of the Next, NHN University. A lot of people were in there.


You can read more details (Korean news) here:
 - http://news.naver.com/main/read.nhn?mode=LSD&mid=sec&sid1=105&oid=138&aid=0001997038

Fault-tolerance in Hama

Recently, Hama core committers Suraj Menon and Thomas Jungblut are working on Fault-tolerant BSP system. And I am trying to read the source code. Their design describes the new BSP computing system and API enabling checkpoint-based recovery. Furthermore, describes the confined recovery, which can be used to improve the cost and latency of recovery.

I didn't fully understand and test yet but quite nice!

PageRank with Apache Hama

The dataset contains 5,716,808 pages and 130,160,392 links and is unzipped ~1gb large.
[oracle@bda03 hama-0.6.0-SNAPSHOT]$ hadoop dfs -ls edward
Found 1 items
-rw-r--r--   3 oracle hadoop 1058414409 2012-08-28 22:16 /user/oracle/edward/links-simple-sorted.txt

[oracle@bda03 hama-0.6.0-SNAPSHOT]$ bin/hama jar hama-examples-0.6.0-SNAPSHOT.jar pagerank /user/oracle/edward/links-simple-sorted.txt out
12/09/10 18:44:25 INFO bsp.FileInputFormat: Total input paths to process : 1
12/09/10 18:44:25 INFO net.NetworkTopology: Adding a new node: /switch1/192.168.10.3:50010
12/09/10 18:44:25 INFO net.NetworkTopology: Adding a new node: /switch1/192.168.10.5:50010
12/09/10 18:44:25 INFO net.NetworkTopology: Adding a new node: /switch1/192.168.10.16:50010
12/09/10 18:44:25 INFO net.NetworkTopology: Adding a new node: /switch1/192.168.10.14:50010
12/09/10 18:44:25 INFO net.NetworkTopology: Adding a new node: /switch1/192.168.10.15:50010
12/09/10 18:44:25 INFO net.NetworkTopology: Adding a new node: /switch1/192.168.10.8:50010
12/09/10 18:44:25 INFO net.NetworkTopology: Adding a new node: /switch1/192.168.10.11:50010
12/09/10 18:44:25 INFO bsp.FileInputFormat: Total # of splits: 17
12/09/10 18:44:25 INFO bsp.BSPJobClient: Running job: job_201209101841_0002
12/09/10 18:44:28 INFO bsp.BSPJobClient: Current supersteps number: 2
12/09/10 18:44:31 INFO bsp.BSPJobClient: Current supersteps number: 8
12/09/10 18:44:34 INFO bsp.BSPJobClient: Current supersteps number: 16
12/09/10 18:44:37 INFO bsp.BSPJobClient: The total number of supersteps: 16
12/09/10 18:44:37 INFO bsp.BSPJobClient: Counters: 10
12/09/10 18:44:37 INFO bsp.BSPJobClient:   org.apache.hama.bsp.JobInProgress$JobCounter
12/09/10 18:44:37 INFO bsp.BSPJobClient:     LAUNCHED_TASKS=17
12/09/10 18:44:37 INFO bsp.BSPJobClient:   org.apache.hama.bsp.BSPPeerImpl$PeerCounter
12/09/10 18:44:37 INFO bsp.BSPJobClient:     SUPERSTEPS=16
12/09/10 18:44:37 INFO bsp.BSPJobClient:     SUPERSTEP_SUM=272
12/09/10 18:44:37 INFO bsp.BSPJobClient:     COMPRESSED_BYTES_SENT=802984424
12/09/10 18:44:37 INFO bsp.BSPJobClient:     TIME_IN_SYNC_MS=13996
12/09/10 18:44:37 INFO bsp.BSPJobClient:     IO_BYTES_READ=1058414409
12/09/10 18:44:37 INFO bsp.BSPJobClient:     COMPRESSED_BYTES_RECEIVED=802984424
12/09/10 18:44:37 INFO bsp.BSPJobClient:     TOTAL_MESSAGES_SENT=11412446
12/09/10 18:44:37 INFO bsp.BSPJobClient:     TASK_INPUT_RECORDS=5706070
12/09/10 18:44:37 INFO bsp.BSPJobClient:     TOTAL_MESSAGES_RECEIVED=5706223
Job Finished in 12.322 seconds