In 2009, Amit Vij and Nima Neghaban founded GIS Federal, a developer of software they called GPUdb. The GIS stood for Global Intelligence Solutions. GPUdb was initially marketed for US military and intelligence applications, at Fort Belvoir for INSCOM.
The United States Postal Service deployed GPUdb in to production 2014.
In 2014 and 2016, the analyst firm International Data Corporation mentioned Kinetica for its production deployments at the US Army and United States Postal Service, respectively. As a result of their work with USPS, IDC announced that Kinetica was the recipient of the HPC Innovation Excellence Award.
On March 3, 2016, the name of the company was changed to GPUdb to match the name of the software, and a $7 million investment was announced which included Raymond J. Lane. In September 2016, it announced another $6 million investment, and an office in San Francisco, while keeping its office in Arlington, Virginia. After adding marketing and service people, the name of both the company and product was changed to Kinetica.
In June 2017, the company announced USD$ 50 million in Series A funding led by Canvas Ventures and Meritech Capital Partners, along with new investor Citi Ventures and existing backer Ray Lane of GreatPoint Ventures
Kinetica makes use GPU to perform equijoins (sort-merge), predicate joins (nested loop), fixed-length string processing, aggression/window function and rendering. Because GPUs are good at handling SIMT (single instruction multiple thread) and simple data structure.
To make use of GPUs more efficiently, Kinetica encourages data locality to minimize data movement from CPU to GPU.
Kinetica would do query plan before query execution to build the optimal query. It can also make use of the users' supplied query hints and existed column indexes to improve the query plan.
Kinetica applys query to each chunk and chunk results are merged hierarchically to get a final result.
Kinetica supports data compression by individual column. Dictionary encoding can be applied to individual columns of restricted-length (charN) type, int type, or long type. During the query execution, when data is retrieved, Kinetica can temporarily decompressed a copy and discard the copy later. For data added or modification, the affected data segment will be uncompressed, modified, and then recompressed immediately.
Nested Loop Join Sort-Merge Join Index Nested Loop Join
Kinetica supports the SQL concept of joining data sets. Since Kinetica makes use of GPU acceleration, tables being joined together must either be replicated or be sharded on the columns being used to join the tables to avoid data communication. Besides, distributed joins, or joins that connect sharded tables on columns other than their shard keys, are not supported.
Join in Kinetican is creating a join view that can be refreshed and future filtering operations.
Kinetica uses primary key index, relation index and column index to improve data access performance.
A primary key index is created by default when a table is created with a primary key specified. The primary key index is hash-based and optimizes the performance of equality-based filter expressions.
A relational index is created as the result of applying a foreign key to a column.
A column index can be applied to a column in a table or view to improve the performance of operations applied to that column in an expression.The column index is implemented as a b-tree, which provides performance improvements for both equality-based and range-based filter criteria on individual columns. Column indexes can also be applied to the primary key columns.
To optimize throughput and delivery fast query process, Kinetica runs completely in-memory. It make use of RAM and VRAM ( the memory for GPU cards). Hot data would be kept in VRAM to optimize data access and avoid data movement between RAM an VRAM. Kinetica needs a warm phase to load data from disk to memory.
https://www.kinetica.com/docs/index.html
Kinetica DB Inc.
2009
GPUdb
C#, C++, Java, JavaScript, Python