Lock-free data structures implement concurrent objects without the use of mutual exclusion. This approach can avoid performance problems due to unpredictable delays while processes are within critical sections. Although universal methods are known that give lock-free data structures for any abstract data type, the overhead of these methods makes them inefficient when compared to conventional techniques using mutual exclusion, such as spin locks. We give lock-free data structures and algorithms for implementing a shared singly-linked list, allowing concurrent traversal, insertion, and deletion by any number of processes. We also show how the basic data structure can be used as a building block for other lock-free data structures. Our algorithms use the single word Compare-and-Swap synchronization primitive to implement the linked list directly, avoiding the overhead of universal methods, and are thus a practical alternative to using spin locks. 1 Introduction A concurrent object is an...

Clusters of multiprocessors, or Clumps, promise to be the supercomputers of the future, but obtaining high performance on these architectures requires an understanding of interactions between the multiple levels of interconnection. In this paper,wepresent the rst multi-protocol implementation of a lightweight message layer|a version of Active Messages-II running on a cluster of Sun Enterprise 5000 servers connected with Myrinet. This research brings together several pieces of high-performance interconnection technology: bus backplanes for symmetric multiprocessors, low-latency networks for connections between machines, and simple, user-level primitives for communication. The paper describes the shared memory message-passing protocol and analyzes the multiprotocol implementation with both microbenchmarks and Split-C applications. Three aspects of the communication layer are critical to performance: the overhead of cache-coherence mechanisms, the method of managing concurrent access, and the cost of accessing state with the slower protocol. Through the use of an adaptive polling strategy, the multi-protocol implementation limits performance interactions between the protocols, delivering up to 160 MB/s of bandwidth with 3.6 microsecond end-to-end latency. Applications within an SMP bene t from this fast communication, running up to 75 % faster than on a network of uniprocessor workstations. Applications running on the entire Clump are limited by the balance of NIC's to processors in our system, and are typically slower than on the NOW. These results illustrate several potential pitfalls for the Clumps architecture. 1

Most multiprocessors are multiprogrammed in order to achieve acceptable response time and to increase their uti-lization. Unfortunately, inopportune preemption may significantly degrade the performance of synchronized parallel applications. To address this problem, researchers have developed two principal strategies for concurrent, atomic update of shared data structures: (1) preemption-safe locking and (2) non-blocking (lock-free) algorithms. Preemption-safe locking requires kernel support. Non-blocking algorithms generally require a universal atomic primitive such as compare-and-swap orload-linked/store-conditional, and are widely regarded as inefficient. We evaluate the performance of preemption-safe lock-based and non-blocking implementations of important data structures—queues, stacks, heaps, and counters—including non-blocking and lock-based queue algorithms of our own, in micro-benchmarks and real applications on a 12-processor SGI Challenge multiprocessor. Our results indicate that our non-blocking queue consistently outperforms the best known alternatives, and that data-structure-specific non-blocking algorithms, which exist for queues, stacks, and counters, can work extremely well. Not only do they outperform preemption-safe lock-based algorithms on multiprogrammed machines, they also outperform ordinary locks on dedicated machines. At the same time, since general-purpose non-blocking techniques do not yet appear to be practical, preemption-safe locks remain the preferred alternative for complex data structures: they outperform

Mutual exclusion locks remain the de facto mechanism for concurrency control on shared-memory data structures. However, their apparent simplicity is deceptive: it is hard to design scalable locking strategies because locks can harbour problems such as priority inversion, deadlock and convoying. Furthermore, scalable lock-based systems are not readily composable when building compound operations. In looking for solutions to these problems, interest has developed in nonblocking systems which have promised scalability and robustness by eschewing mutual exclusion while still ensuring safety. However, existing techniques for building non-blocking systems are rarely suitable for practical use, imposing substantial storage overheads, serialising non-conflicting operations, or requiring instructions not readily available on today’s CPUs. In this paper we present three APIs which make it easier to develop non-blocking implementations of arbitrary data structures. The first API is a multi-word compare-and-swap operation (MCAS) which atomically updates a set of memory locations. This can be used to advance a data structure from one consistent state to another. The second API is a word-based software transactional memory (WSTM) which can allow sequential code to be re-used more directly than with MCAS and which provides better scalability when locations are being read rather than being

High performance parallel programs are currently difficult to write and debug. One major source of difficulty is protecting concurrent accesses to shared data with an appropriate synchronization mechanism. Locks are the most common mechanism but they have a number of disadvantages, including possibly unnecessary serialization, and possible deadlock. Transactional memory is an alternative mechanism that makes parallel programming easier. With transactional memory, a transaction provides atomic and serializable operations on an arbitrary set of memory locations. When a transaction commits, all operations within the transaction become visible to other threads. When it aborts, all operations in the transaction are rolled back. Transactional

The memory consistency model for a shared-memory multiprocessor specifies the behavior of memory with respect to read and write operations from multiple processors. As such, the memory model influences many aspects of system design, including the design of programming languages, compilers, and the underlying hardware. Relaxed models that impose fewer memory ordering constraints offer the potential for higher performance by allowing hardware and software to overlap and reorder memory operations. However, fewer ordering guarantees can compromise programmability and portability. Many of the previously proposed models either fail to provide reasonable programming semantics or are biased toward programming ease at the cost of sacrificing performance. Furthermore, the lack of consensus on an acceptable model hinders software portability across different systems. This dissertation focuses on providing a balanced solution that directly addresses the trade-off between programming ease and performance. To address programmability, we propose an alternative method for specifying memory behavior that presents a higher level abstraction to the programmer. We show that with only a few types of information supplied by the

Programs written in concurrent object-oriented languages, espe-cially ones that employ thread-safe reusable class libraries, can execute synchronization operations (lock, notify, etc.) at an amaz-ing rate. Unless implemented with utmost care, synchronization can become a performance bottleneck. Furthermore, in languages where every object may have its own monitor, per-object space overhead must be minimized. To address these concerns, we have developed a meta-lock to mediate access to synchronization data. The meta-lock is fast (lock + unlock executes in 11 SPARCTM architecture instructions), compact (uses only two bits of space), robust under contention (no busy-waiting), and flexible (supports a variety of higher-level synchronization operations). We have vali-dated the meta-lock with an implementation of the synchronization operations in a high-performance product-quality JavaTM virtual machine and report performance data for several large programs.

Work on non-blocking data structures has proposed extending processor designs with a compare-and-swap primitive, CAS2, which acts on two arbitrary memory locations. Experience suggested that current operations, typically single-word compare-and-swap (CAS1), are not expressive enough to be used alone in an efficient manner. In this paper we build CAS2 from CAS1 and, in fact, build an arbitrary multi-word compare-and-swap (CASN). Our design requires only the primitives available on contemporary systems, reserves a small and constant amount of space in each word updated (either 0 or 2 bits) and permits nonoverlapping updates to occur concurrently. This provides compelling evidence that current primitives are not only universal in the theoretical sense introduced by Herlihy, but are also universal in their use as foundations for practical algorithms. This provides a straightforward mechanism for deploying many of the interesting non-blocking data structures presented in the literature that have previously required CAS2.

We introduce the concept of Hierarchical Clustering as a way to structure shared memory multiprocessor operating systems for scalability. As the name implies, the concept is based on clustering and hierarchical system design. Hierarchical Clustering leads to a modular system, composed of easy-todesign and efficient building blocks. The resulting structure is scalable because it i) maximizes locality, which is key to good performance in NUMA systems, and ii) provides for concurrency that increases linearly with the number of processors. At the same time, there is tight coupling within a cluster, so the system performs well for local interactions which are expected to constitute the common case. A clustered system can easily be adapted to different hardware configurations and architectures by changing the size of the clusters. We show how this structuring technique is applied to the design of a microkernel-based operating system called HURRICANE. This prototype system is the first complete and running implementation of its kind, and demonstrates the feasibility of a hierarchically clustered system. We present performance results based on the prototype, demonstrating the characteristics and behavior of a clustered system. In particular, we show how clustering trades off the efficiencies of tight coupling for the advantages of replication, increased locality, and decreased lock contention. We describe some of the lessons we learned from our implementation efforts and close with a discussion of our future work.

Most complexitymeasures for concurrent algorithms for asynchronous sharedmemory architectures focus on process steps and memory consumption. In practice, however, performance of multiprocessor algorithms is heavily influenced by contention, the extent to which processes access the same location at the same time. Nevertheless, even though contention is one of the principal considerations affecting the performance of real algorithms on real multiprocessors, there are no formal tools for analyzing the contention of asynchronous shared-memory algorithms. This paper introduces the first formal complexity model for contention in multiprocessors. We focus on the standard multiprocessor architecture in which n asynchronous processes communicate by applying read, write, and read-modify-write operations to a shared memory. We use our model to derive two kinds of results: (1) lower bounds on contention for well known basic problems such as agreement and mutual exclusion, and (2) trade-offs betwe...