This paper examines simultaneous multithreading, a technique permitting several independent threads to issue instructions to a superscalar’s multiple functional units in a single cycle. We present several models of simultaneous multithreading and compare them with alternative organizations: a wide superscalar, a fine-grain multithreaded processor, and single-chip, multiple-issue multiprocessing architectures. Our results show that both (single-threaded) superscalar and fine-grain multithreaded architectures are limited in their ability to utilize the resources of a wide-issue processor. Simultaneous multithreading has the potential to achieve 4 times the throughput of a superscalar, and double that of fine-grain multithreading. We evaluate several cache configurations made possible by this type of organization and evaluate tradeoffs between them. We also show that simultaneous multithreading is an attractive alternative to single-chip multiprocessors; simultaneous multithreaded processors with a variety of organizations outperform corresponding conventional multiprocessors with similar execution resources. While simultaneous multithreading has excellent potential to increase processor utilization, it can add substantial complexity to the design. We examine many of these complexities and evaluate alternative organizations in the design space.

The large latency of memory accesses is a major obstacle in obtaining high processor utilization in large scale shared-memory multiprocessors. Although the provision of coherent caches in many recent machines has alleviated the problem somewhat, cache misses still occur frequently enough that they significantly lower performance. In this paper we evaluate the effectiveness of non-binding software-controlled lyrefetching, as proposed in the Stanford DASH Multiprocessor, to address this problem. The prefetches are non-binding in the sense that the prefetched data is brought to a cache close to the processor, but is still available to the cache coherence protocol to keep it consistent. Prefetching is software-controlled since the program must explicitly issue prefetch instructions.

Non-blocking caches and prefetching caches are two techniques for hiding memory latency by exploiting the overlap of processor computations with data accesses. A non-blocking cache allows execution to proceed concurrently with cache misses as long as dependency constraints are observed, thus exploiting post-miss operations. A prefetching cache generates prefetch requests to bring data in the cache before it is actually needed, thus allowing overlap with pre-miss computations. In this paper, we evaluate the effectiveness of these two hardware-based schemes. We propose a hybrid design based on the combination of these approaches. We also consider compiler-based optimizations to enhance the effectiveness of non-blocking caches. Results from instruction level simulations on the SPEC benchmarks show that the hardware prefetching caches generally outperform non-blocking caches. Also, the relative effectiveness of non-blocking caches is more adversely affected by an increase in memory latency...

Dataflow architectures tolerate long unpredictable com-munication delays and support generation and coordi-nation of parallel activities directly in hardware, rather than assuming that program mapping will cause these issues to disappear. However, the proposed mecha-nisms are complex and introduce new mapping com-plications. This paper presents a greatly simplified ap-proach to dataflow execution, called the explicit token store (ETS) architecture, and its current realization in Monsoon. The essence of dynamic datallow execution is captured by a simple transition on state bits associ-ated with storage local to a processor. Low-level storage management is performed by the compiler in assigning nodes to slots in an activation frame, rather than dy-namically in hardware. The processor is simple, highly pipelined, and quite general. It may be viewed as a generalization of a fairly primitive von Neumann archi-tecture. Although the addressing capability is restric-tive, there is exactly one instruction executed for each action on the dataflow graph. Thus, the machine ori-ented ETS model provides new understanding of the merits and the real cost of direct execution of dataflow graphs. 1

The Alewife multiprocessor project focuses on the architecture and design of a large-scale parallel machine. The machine uses a low-dimensional direct interconnection network to provide scalable communication bandwidth, while allowing the exploitation of locality. Despite its distributed-memory architecture, Alewife allows efficient shared-memory programming through a multilayered approach to locality management. A new scalable cache-coherence scheme called LimitLESS directories allows the use of caches for reducing communication latency and network bandwidth requirements. Alewife also employs run-time and compile-time methods for partitioning and placement of data and processes to enhance communication locality. While the above methods attempt to minimize communication latency, communication with distant processors cannot be completely avoided. Alewife's processor, Sparcle, is designed to tolerate these latencies by rapidly switching between threads of computation. This paper describe...

Abstract: In this paper, we present a relatively primitive execution model for ne-grain parallelism, in which all synchronization, scheduling, and storage management is explicit and under compiler control. This is de ned by a threaded abstract machine (TAM) with a multilevel scheduling hierarchy. Considerable temporal locality of logically related threads is demonstrated, providing an avenue for e ective register use under quasi-dynamic scheduling. A prototype TAM instruction set, TL0, has been developed, along with a translator to a variety of existing sequential and parallel machines. Compilation of Id, an extended functional language requiring ne-grain synchronization, under this model yields performance approaching that of conventional languages on current uniprocessors. Measurements suggest that the net cost of synchronization on conventional multiprocessors can be reduced to within a small factor of that on machines with elaborate hardware support, such as proposed data ow architectures. This brings into question whether tolerance to latency and inexpensive synchronization require speci c hardware support or merely an appropriate compilation strategy and program representation. 1

This article explores parallel processing on an alternative architecture, simultaneous multithreading (SMT), which allows multiple threads to compete for and share all of the processor's resources every cycle. The most compelling reason for running parallel applications on an SMT processor is its ability to use thread-level parallelism and instruction-level parallelism interchangeably. By permitting This research was supported by Digital Equipment Corporation, the Washington Technology Center, NSF PYI Award MIP-9058439, NSF grants MIP-9632977, CCR-9200832, and CCR9632769, DARPA grant F30602-97-2-0226, ONR grants N00014-92-J-1395 and N00014-94-11136, and fellowships from Intel and the Computer Measurement Group.

... utilization. By maintaining multiple process contexts in hardware and switching among them in a few cycles, multithreaded processors can overlap computation with memory accesses and reduce processor idle time. This paper presents an analytical performance model for multithreaded processors that includes cache interference, network contention, context-switching overhead, and data-sharing effects. The model is validated through our own simulations and by comparison with previously published simulation results. Our results indicate that processors can substantially benefit from multithreading, even in systems with small caches. Large caches yield close to full processor utilization with as few as two to four contexts, while small caches may require up to four times as many contexts. Increased network contention due to multithreading has a major effect on performance. The available network bandwidth and the context-switching overhead limits the best possible utilization.

Techniques that can cope with the large latency of memory accesses are essential for achieving high processor utilization in large-scale shared-memory multiprocessors. In this paper, we consider four architectural techniques that address the latency problem: (i) hardware coherent caches, (ii) relaxed memory consistency, (iii) softwarecontrolled prefetching, and (iv) multiple-context support. While some studies of benefits of the individual techniques have been done, no study evaluates all of the techniques within a consistent framework. This paper attempts to remedy this by providing a comprehensive evaluation of the benefits of the four techniques, both individually and in combinations, using a consistent set of architectural assumptions. The results in this paper have been obtained using detailed simulations of a large-scale shared-memory multiprocessor. Our results show that caches and relaxed consistency uniformly improve performance. The improvements due to prefetching and multiple contexts are sizeable, but are much more applicationdependent. Combinations of the various techniques generally attain better performance than each one on its own. Overall, we show that using suitable combinations of the techniques, performance can be improved by 4 to 7 times.

The M-Machine is an experimental multicomputer being developed to test architectural concepts motivated by the constraints of modern semiconductor technology and the demands of programming systems. The M-Machine computing nodes are con- nected with a 3-D mesh network; each node is a multithreaded processor incorporating 12 function units, on-chip cache, and local memory. The multiple function units are used to exploit both instruction-level and thread-level parallelism. A user accessible message passing system yields fast communication and synchronization between nodes. RapM access to remote memory is provided transparently to the user with a combination of hardware and software mechanisms. This paper presents the architecture of the M-Machine and describes how its mechanisms attempt to maximize both single thread performance and overall system throughput. The architecture is complete and the MAP chip, which will serve as the M-Machine processing node, is currently being implemented.