M. M. Michael and M. L. Scott. Nonblocking algorithms and preemption-safe locking on multiprogrammed shared memory multiprocessors. Journal of Parallel and Distributed Computing, 51:1--26, 1998.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....the aim to lower the contention when the system is in a high congestion situation. These implementations give di erent execution times under di erent contention instances. But still the time spend by the processes on the synchronisation can form a substantial part of the program execution time [9, 15, 16, 18, 29]. The reason for this is that typical synchronisation is based on blocking that introduces performance bottlenecks because of busywaiting and convoying. Busy waiting tends to produce a large amount of memory and interconnection network contention. The convoying e ect that takes place when a ....

....of their synchronisation power [7] Some evaluation studies have also been performed for speci c data structure implementations. Most of these performance evaluations were using micro benchmarks and were performed on small scale symmetric multiprocessors, as well as distributed memory machines [3, 8, 10, 11, 16] or simulators [10, 13] Micro benchmarks are useful since they enable easy isolation of performance issues, but the real goal of better synchronisation methods is to improve performance of real applications, which micro benchmarks may not represent well. A substantial number of realistic scalable ....

M. M. Michael and M. L. Scott, Nonblocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors, Journal of Parallel and Distributed Computing 51(1), pp. 1-26, 1998.

....application failure. Locking algorithms can be modified to address these limitations but require significant effort and creativity from the programmer. Software proposals have been made to make lock based critical sections non blocking [163] and thread scheduling that is aware of blocking locks [87, 123]. Bohannon et al. present a recovery strategy to allow a system to recover from a lock held by a process believed to have failed [20] Locks also restrict parallelism: two operations on the same object cannot execute in parallel, even if they access disjoint parts of the object. In some cases, ....

Maged M. Michael and Michael L. Scott. Nonblocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors. Journal of Parallel and Distributed Computing, 51(1):1--26, 1998.

....interrupt handler. It increases interrupt latency when the interrupt occurred inside the atomic procedure, because the procedure must first be finished. But the most severe downside of this technique is, that it inhibits inlining of memory accesses. Similar techniques are described in [9] 36] [35], 41] Figure 13a shows the change in performance when no revocation checks are performed. This configuration is slightly slower than a configuration that used the CLI method for revocation check. We can only explain this by code cache effects. Using spinlocks adds an additional overhead ....

M. Michael and M. Scott. Nonblocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors. In Journal of Parallel and Distributed Computing, 54(2), pp. 162-182, 1998.

....requires lock acquisitions in the presence of conflicts. Improving performance of software non blocking schemes have been studied previously [27, 4, 38] Software proposals have been made to make lock based critical sections non blocking [37] and thread scheduling that is aware of blocking locks [18, 28]. Database concurrency control. Transactions are well understood and studied in database literature [10] The use of timestamps for resolving conflicts and ordering transactions in database systems has been well studied [5, 32] Optimistic concurrency control (OCC) was proposed as an alternative ....

M. M. Michael and M. L. Scott. Nonblocking algorithms and preemption -safe locking on multiprogrammed shared memory multiprocessors. Journal of Parallel and Distributed Computing, 51(1):1--26, 1998.

....solutions that introduce blocking are penalised by locking that introduces priority inversion, deadlock scenarios and performance bottlenecks. The time that a process can spend blocked while waiting to get access to the critical section can form a substantial part of the algorithm execution time [5, 9, 10, 14]. There are two main reasons that locking is so expensive. The rst reason is the convoying eoeect that blocking synchronisation suoeers from: if a process holding the lock is preempted, any other process waiting for the lock is unable to perform any useful work until the process that hold the ....

....the double compare andswap atomic primitive that is available only on some members of the Motorola 68000 family of processors. Valois in [12] presents a non blocking queue algorithm together with several other non blocking data structures, his queue is an array based one. Michael and Scott in [10] presented a nonblocking queue based on a singly link list, which is the most eOEcient and scalable non blocking algorithm compared with the other algorithms mentioned above. The remainder of the paper is organised as follows. In Section 2 we give a brief introduction to shared memory ....

[Article contains additional citation context not shown here]

M. M. Michael and M. L. Scott. Nonblocking algorithms and preemption-safe locking on multiprogrammed shared memory multiprocessors. Journal of Parallel and Distributed Computing, 51(1):126, 25 May 1998.

....the aim to lower the contention when the system is in a high congestion situation. These implementations give dioeerent execution times under dioeerent contention instances. But still the time spend by the processes on the synchronisation can form a substantial part of the program execution time [9, 15, 16, 18, 29]. The reason for this is that typical synchronisation is based on blocking that introduces performance bottlenecks because of busy waiting and convoying. Busy waiting tends to produce a large amount of memory and interconnection network contention. The convoying eoeect that takes place when a ....

....of their synchronisation power [7] Some evaluation studies have also been performed for specic data structure implementations. Most of these performance evaluations were using micro benchmarks and were performed on small scale symmetric multiprocessors, as well as distributed memory machines [3, 10, 11, 8, 16] or simulators [10, 13] Micro benchmarks are useful since they enable easy isolation of performance issues, but the real goal of better synchronisation methods is to improve performance of real applications, which micro benchmarks may not represent well. A substantial number of realistic scalable ....

M. M. Michael and M. L. Scott, Nonblocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors, Journal of Parallel and Distributed Computing 51(1), pp. 1-26, 1998.

....solutions that introduce blocking are penalised by locking that introduces priority inversion, deadlock scenarios and performance bottlenecks. The time that a process can spend blocked while waiting to get access to the critical section can form a substantial part of the algorithm execution time [7, 11, 12, 18]. There are two main reasons that locking is so expensive. The rst reason is the convoying eoeect that blocking synchronisation suoeers from: if a process holding the lock is preempted, any other process waiting for the lock is unable to perform any useful work until the process that hold the ....

....the double compare and swap atomic primitive that is available only on some members of the Motorola 68000 family of processors. Valois in [16] presents a non blocking queue algorithm together with several other non blocking data structures, his queue is an array based one. Michael and Scott in [12] presented a nonblocking queue based on a singly link list, which is the most eOEcient and scalable non blocking algorithm compared with the other algorithms mentioned above. The remainder of the paper is organised as follows. In Section 2 we give a brief introduction to shared memory ....

[Article contains additional citation context not shown here]

M. M. Michael and M. L. Scott, Nonblocking Algorithms and PreemptionSafe Locking on Multiprogrammed Shared Memory Multiprocessors, Journal of Parallel and Distributed Computing 51(1), pp. 1-26, 1998.

....solutions that introduce blocking are penalised by locking that introduces priority inversion, deadlock scenarios and performance bottlenecks. The time that a process can spend blocked while waiting to get access to the critical section can form a substantial part of the algorithm execution time [7, 11, 12, 18]. There are two main reasons that locking is so expensive. The rst reason is the convoying eoeect that blocking synchronisation suoeers from: if a process holding the lock is preempted, any other process waiting for the lock is unable to perform any useful work until the process that hold the ....

....the doublecompareandswap atomic primitive that is available only on some members of the Motorola 68000 family of processors. Valois in [16] presents a non blocking queue algorithm together with several other non blocking data structures, his queue is an array based one. Michael and Scott in [12] presented a nonblocking queue based on a singly link list, which is the most eOEcient and scalable non blocking algorithm compared with the other algorithms mentioned above. The remainder of the paper is organised as follows. In Section 2 we give a brief introduction to shared memory ....

[Article contains additional citation context not shown here]

M. M. Michael and M. L. Scott, Nonblocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors, Journal of Parallel and Distributed Computing 51(1), pp. 1-26, 1998.

....implementations that lower the contention when the system is in a high congestion situation, and they give dioeerent execution times under dioeerent contention instances. But still the time spend by the processes on the synchronisation can form a substantial part of the program execution time [8, 14, 15, 17, 27]. The reason for this is that in any shared memory parallel systems concurrent processes that have been created either by a parallel application or by the operating system very often need to share data or become coordinated, and they do so via shared data objects structures. To ensure consistency ....

....Some studies have been focused on the developing of better software algorithms, while others have identied the properties of dioeerent atomic transaction operations in terms of their synchronisation power. Some evaluation studies have also been performed for specic data structure implementations [15]. Most of these performance evaluations were using micro benchmarks and were performed on small scale symmetric multiprocessors, as well as distributed memory machines [3, 9, 10, 7] or simulators [9, 12] Micro benchmarks are useful since they enable easy isolation of performance issues, but the ....

M. M. Michael and M. L. Scott, Nonblocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors, Journal of Parallel and Distributed Computing 51(1), pp. 1-26, 1998.

.... studies with different benchmark suites and programming models [8, 19] exemplify that the overhead of synchronization remains a dominant bottleneck in parallel programs for cache coherent shared memory multiprocessors, despite the enormousness of related work that appears in the literature [1, 4, 6, 7, 9, 11, 14, 16]. 1 To appear in the IEEE 2000 International Parallel and Distributed Processing Symposium, Cancun, Mexico, May 1 5 2000. Synchronization operations account frequently for significant fractions of execution time and may limit the scalability of parallel programs on large processor scales and ....

.... not appeared in the literature until recently [10, 17] Lock free synchronization has also attracted considerable attention due to its competitive performance compared to lock based synchronization and its robustness as a synchronization discipline in multiprogrammed shared memory multiprocessors [2, 13, 16, 17, 20]. Synchronization primitives on shared memory multiprocessors can be analyzed effectively through time decomposition of synchronization periods [5] A generic synchronization primitive can be decomposed into at most four distinct time intervals, the acquire, the waiting, the compute and the ....

M. Michael and M. Scott. Non-blocking Algorithms and Preemption-safe Locking on Multiprogrammed Shared Memory Multiprocessors. Journal of Parallel and Distributed Computing, 54(2), pp. 162--182, Feb. 1998.

....to avoid introducing hot spots in the memory system. A strong advan tage of non blocking queues and memory pools is that the associated synchronization operations are immune to inopportune preemptions of threads from the operating system and therefore robust in the presence of multiprogramming [9]. The key feature that enables multiprogramming scalability of nanothreaded programs is a shared arena [3, 12] used to export kernel side processor allocation decisions to the program, as well as the state of kernel threads that serve as the execution vehicles of the program. This information is ....

M. Michael and M. Scott. Nonblocking Algorithms and Preemption-Safe Locking in Multiprogrammed Shared Memory Multiprocessors. Journal of Parallel and Distributed Computing, 54(2), pp. 162--182, 1998.

....kernel exports a minimal set of services to the user level, which are exploited from the runtime system to implement dynamic process control within each parallel program. The process control mechanisms are coupled with a dynamic user level scheduler [14] and non blocking synchronization [10]. The current implementation is integrated with a user level threads library which uses compiler knowledge to control threads granularity and match the number of running user level threads of a parallel program to the number of EVs granted to the program by the kernel [3, 4] However, the ....

M. Michael and M. Scott, Nonblocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors, Journal of Parallel and Distributed Computing, Vol. 54(2), pp. 162--182, 1998.

No context found.

M. M. Michael and M. L. Scott. Nonblocking algorithms and preemption-safe locking on multiprogrammed shared memory multiprocessors. Journal of Parallel and Distributed Computing, 51:1--26, 1998.

No context found.

Maged M. Michael and Michael L. Scott. Nonblocking algorithms and preemption-safe locking on multiprogrammed shared --- memory multiprocessors. Journal of Parallel and Distributed Computing, 51(1):1--26, 1998.

No context found.

M. M. Michael and M. L. Scott. Nonblocking algorithms and preemption-safe locking on multiprogrammed shared --- memory multiprocessors. Journal of Parallel and Distributed Computing, 51(1):1--26, 1998.

No context found.

Michael, M. and Scott, M. Non-Blocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors. Journal of Parallel and Distributed Computing, 1998.

No context found.

Michael M., Scott M. Non-Blocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors. Journal of Parallel and Distributed Computing. 1998.

No context found.

Michael M., Scott M. Non-Blocking Algorithms and Preemption-Safe Locking on Multiprogrammed Shared Memory Multiprocessors. Journal of Parallel and Distributed Computing. 1998.

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