Ruoming Jin and Gagan Agrawal. Shared Memory Parallelization of Data Mining Algorithms: Techniques, Programming Interface, and Performance. In Proceedings of the second SIAM conference on Data Mining, April 2002. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Compiler and Runtime Support for Shared Memory.. - Li, Jin, Agrawal (2002) Self-citation (Jin Agrawal) (Correct)
No context found.
Ruoming Jin and Gagan Agrawal. Shared Memory Parallelization of Data Mining Algorithms: Techniques, Programming Interface, and Performance. In Proceedings of the second SIAM conference on Data Mining, April 2002.
Communication and Memory Efficient Parallel Decision Tree.. - Jin, Agrawal (2003) Self-citation (Jin Agrawal) (Correct)
No context found.
Ruoming Jin and Gagan Agrawal. Shared Memory Parallelization of Data Mining Algorithms: Techniques, Programming Interface, and Performance. In Proceedings of the second SIAM conference on Data Mining, April 2002.
Performance Prediction for Random Write Reductions: A Case.. - Jin, Agrawal (2002) (1 citation) Self-citation (Jin Agrawal) (Correct)
....vendors. Vendors of these machines are targeting data warehousing and data mining as major markets. Thus, shared memory parallelization of random write reductions is of significant interest. In our previous work, we have developed several techniques for parallelizing random write reductions [10, 11]. One of the tech niques involves creating a copy of the reduction object for each thread and is referred to as full replication. The other techniques use locking to avoid race conditions. Among the locking schemes, two have shown particularly promising performance. They are optimized full ....
....element in the reduction object. After processing a data item, a thread needs to acquire the lock associated with the element in the reduction object it needs to update. In our experiment with apriori, with 2000 distinct items and support level of 0. 1 , up to 3 million candidates were generated [11]. In full locking, this means supporting 3 million locks. Supporting such a large numbers of locks results in overheads of three types. The first is the high memory requirement associated with a large number of locks. The second overhead comes from cache misses. Consider an update operation. If ....
[Article contains additional citation context not shown here]
Ruoming Jin and Gagan Agrawal. Shared Memory Parallelization of Data Mining Algorithms: Techniques, Programming Interface, and Performance. In Proceedings of the second SIAM conference on Data Mining, April 2002.
Distributed Data Mining Bibliography - Hillol (Correct)
No context found.
R. Jin and G. Agrawal. Shared Memory Parallelization of Data Mining Algorithms: Techniques, Programming Interface, and Performance. In Proceedings of the Second SIAM International Conference on Data Mining, Arlington, VA, April 2002.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC