diff options
| author | msabhi <abhi.is2006@gmail.com> | 2016-11-17 16:41:41 -0500 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2016-11-17 16:41:41 -0500 |
| commit | d63ceb8aa8c280d789e3163426ea86bff16997e0 (patch) | |
| tree | baaafed7b33831d663d4dd36d14a80956f96a458 | |
| parent | 99b7aadcaf75b04c9d7a31d6d9671d13f609db36 (diff) | |
Update big-data.md
| -rw-r--r-- | chapter/8/big-data.md | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/chapter/8/big-data.md b/chapter/8/big-data.md index 9dfd1d6..6a3edd3 100644 --- a/chapter/8/big-data.md +++ b/chapter/8/big-data.md @@ -14,6 +14,7 @@ In BSP model + The processors involved have their own local memory and every processor is connected to other via a point-to-point communication. + At every superstep, a processor receives input at the beginning, performs computation and outputs at the end. + Barrier synchronization synchs all the processors at the end of every superstep. ++ A notable feature of the model is the complete control on data through communication between every processor at every superstep. <br /> Though similar to map reduce model, BSP preserves data in memory across supersteps and helps in reasoning iterative graph algorithms.<br /> `Pregel` is an implementation of classic BSP model by Google (PageRank) to analyze large graphs exclusively. It was followed by open source implementations - Apache’s Giraph and Hama; which were BSP models built on top of Hadoop. |