Skip to content


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
GPU database engine
Cuda C C++ Bison
branch: master
Failed to load latest commit information. bug fixes
Makefile modified the Makefile A few updates . A few updates Modified load scripts Bitmap index support & SSD storage support
alenka.h Exposed data.dict Added Makefiles for Windows and Linux
atof.h Added Makefiles for Windows and Linux Minor changes Minor changes Minor changes
bison.y Minor updates
callbacks.c License change
callbacks.h License change
change.log Changed the storage model for strings Better GroupBy performance
cm.h Better GroupBy performance Better GroupBy performance Code cleanup
filter.h License change
fl.l Minor changes change
how_to_run_tpch.txt Added how_to_run_tpch file
lex.yy.c Minor changes Replaced "REFERENCES" with run-time checks Database load modifications
manual.txt Added better formatting. Better performance for group operator
merge.h License change updated Minor changes
operators.h Moved most ops to . Better GroupBy performance
select.h Changed the license to GPL 3.0 Performance improvements for group operator Bitmap index support & SSD storage support
stdint.h 4-4-2012-commit Minor changes
strings.h Performance improvements for group operator Scalability-related changes License change License change
strings_type.h Bitmap index support & SSD storage support A few fixes Minor updates
zone_map.h License change

Welcome to Alenka - GPU database engine

How to build?

Download Alenka

Download ModernGpu library

Unzip ModernGPU into Alenka directory

Run Makefile or

What is this?

This is a GPU based database engine written to use vector based processing and high bandwidth of modern GPUs

Features :

  • Vector-based processing
    CUDA programming model allows a single operation to be applied to an entire set of data at once.

  • Smart compression
    Ultra fast compression and decompression on GPU. Database operations on compressed data.

  • Column-based storage
    Minimizes disk I/O by only accessing the relevant data.

  • Data skipping
    Better performance without indexes.

  • Fast Loading
    Gpu based CSV parser loads the data into database at very high speed.

How to use it ?

Create your data files :

Run scripts load_orders.sql, load_lineitem.sql and load_customer.sql to create your database files.

Run your queries from a command prompt or use Alenka JDBC driver from Technica Corporation

Step 1 - Filter data

OFI := FILTER orders BY o_orderdate < 19950315;

CF := FILTER customers BY c_mktsegment == "BUILDING";

LF := FILTER lineitem BY shipdate > 19950315;

Step 2 - Join data

OLC := SELECT o_orderkey AS o_orderkey, o_orderdate AS o_orderdate, o_shippriority AS o_shippriority, price AS price, discount AS discount FROM LF JOIN OFI ON orderkey = o_orderkey JOIN CF ON o_custkey = c_custkey;

Step 3 - Group data

F := SELECT o_orderkey AS o_orderkey1, o_orderdate AS orderdate1, o_shippriority AS priority, SUM(price*(1-discount)) AS sum_revenue, COUNT(o_orderkey) AS cnt
FROM OLC GROUP BY o_orderkey, o_orderdate, o_shippriority;

Step 4 - Order data

RES := ORDER F BY sum_revenue DESC, orderdate1 ASC;

Step 5 - Save the results

STORE RES INTO 'results.txt' USING ('|') LIMIT 10;

Something went wrong with that request. Please try again.