Bleu Framework for Hbase & Hadoop

Presentation for Cloud-based applications

Transcript

1. Bleu Framework for Hbase & Hadoop 49941131 JIA-HUEI REN Bleu.tw@gmail.com
2. NoSQL, Not Only SQL select fun, profit from real_world where relational=false;
3. Behavior • Increase the machine will automatically expand the data storage. • The schema-less format of a Key-Value. • CAP – Consistent – Availability – Partition Tolerance
4. Consistent • All nodes see the same data at the same time.
5. Availability • A guarantee that every request receives a response about whether it was successful or failed.
6. Partition Tolerance • The system continues to operate despite arbitrary message loss or failure of part of the system.
7. Why ?
8. Recording?
9. Recording?
10. Analyzing
11. Visualizing
12. Database Transfer
13. Usage • Big Users • Big Data • Internet of Things • Cloud Computing
14. So…I make a framework For NoSQL and RDBMS
15. Bleu Framework • High Compatibility • High Transferability • Low coupling • High cohesion
16. What’s MVC
17. What’s Thrift PHP Hbase Thrift Hadoop
18. Thrift Struct Exception Function struct TCell struct ColumnDescriptor struct TRegionInfo struct Mutation struct BatchMutation struct TIncrement struct TColumn struct TRowResult struct TScan exception IOError exception IllegalArgument exception AlreadyExists enableTable() disableTable() bool isTableEnabled() getTableNames() getColumnDescriptors() createTable() deleteTable() get() getRow() getRowWithColumns() getRows() mutateRow() mutateRows() deleteAll() deleteAllRow() scannerOpen() scannerGet() scannerClose()
19. Create & Drop table Schema::create('users', function($table) { $table->increments('id'); } ); Schema::drop('users');
20. Get • getTableNames • getTable • getTables
21. Select User ::all(); User ::find(1);
22. Insert $user = new User; $user->name = 'John'; $user->save();
23. Update $user = User::find(1); $user->email = 'john@foo.com'; $user->save();
24. Delete $user = User::find(1); $user->delete(); User::destroy(array(1, 2, 3));
25. Performance Comparison 0 50 100 150 200 250 300 350 400 450 500 1000 20000 100000 1000000 2000000 Hadoop MySQL
26. Conclusions • Use Thrift-PHP will reduce the performance of Hbase. • This framework can easily switch between NoSQL database and RDBMS. • ORM design gives this framework more readable.
27. Further works • More database engine support. • Performance • Flexible
28. Show Time :)
29. Thanks!