Between Dec 19, 2024 and Jan 2, 2025, datasets can be submitted to DRUM but will not be processed until after the break. Staff will not be available to answer email during this period, and will not be able to provide DOIs until after Jan 2. If you are in need of a DOI during this period, consider Dryad or OpenICPSR. Submission responses to the UDC may also be delayed during this time.

Browsing by Author "Kong, Jinseok"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A Non-blocking Directory Protocol for Large-Scale Multiprocessors
    (1999-04-08) Kong, Jinseok; Yew, Pen-Chung; Lee, Gyungho
    This paper presents a non-blocking directory-based cache coherence protocol to improve the performance of large-scale distributed shared-memory multiprocessors. In the proposed non-blocking directory protocol, all subsequent request can proceed without being blocked at the directory. The critical path of a data access and the amount of network transactions needed to complete a memory request (including the transactions needed to maintain coherence) are also reduced. To support the non-blocking directory protocol, the history of data accesses is maintained with a small history table at the directory. Using detailed simulations,we evaluate the performance of the protocol compared with other blocking directory protocols.
  • Thumbnail Image
    Item
    Minimizing the Directory Size for Large-scale DSM Multiprocessors
    (1999-01-09) Kong, Jinseok; Yew, Pen-Chung; Lee, Gyungho
    Directory-based cache coherence schemes are commonly used in large-scale distributed shared-memory (DSM) multiprocessors, but most of them rely on heuristics to avoid large hardware requirements. We proposed to use physical address mapping on the memory modules to significantly reduce the directory size needed. This approach allows the size of directory to grow as O (cn log2n) as in optimal pointer-based directory schemes [10], where n is the number of nodes in the system and c is the number of cache lines in each cache memory. Our simulations show that the proposed scheme can achieve a performance equivalent to the heuristic pointer-based directory schemes, but with a smaller directory size and a simpler pointer management scheme.