We introduce the concept of deadlock-free migration-based coherent shared memory to the NUCA family of architectures. Migration-based architectures move threads among cores to guarantee sequential semantics in large multicores. Using a execution migration (EM) architecture, we achieve performance comparable to directory-based architectures without using directories: avoiding automatic data replication significantly reduces cache miss rates, while a fast network-level thread migration scheme takes advantage of shared data locality to reduce remote cache accesses that limit traditional NUCA performance. EM area and energy consumption are very competitive, and, on the average, it outperforms a directory-based MOESI baseline by 6.8% and a traditional S-NUCA design by 9.2%. We argue that with EM scaling performance has much lower cost and design complexity than in directorybased coherence and traditional NUCA architectures: by merely scaling network bandwidth from 128 to 256 (512) bit flits, the performance of our architecture improves by an additional 8% (12%), while the baselines show negligible improvement.
In the context of processing data lists in parallel in a multicore platform, various threads share a workload, each using a list to get and insert the data items to be processed; and when a list becomes empty, the owner thread steals data items from another list — thus balancing the workload according to the processing capacity of each thread and transparently to the programmer.
In recent years search for proper extensions of the object model to suit real-time database community has become a critical research issue. In this paper, we present such an extension of the object model, called RTO-RTDB. A concrete static and dynamic views of the model are presented. A major attraction of this model is in that it is effective not only in the use of the feedback control real-time scheduling theory but also in the accurate representation of realtime databases properties. It is used to demonstrate how to separate structural and concurrency concerns ensuring also high-level abstraction for parallelism modeling. In order to improve and to facilitate the RTO-RTDB object model specification, we have used the MARTE and UML-RTDB profiles as UML extensions to describe real-time aspects on one hand, and to express real-time database features on the other hand.
Speculative multithreading is one of the most promising methods for speeding up the execution of programs in multicore systems. Each loop has numerous possible execution paths; however, in many cas...
Today there exist many programming models and platforms for implementing real-time stream processing systems. A decision in favor of the wrong technology might lead to increased development time an ...
With the proliferation of multiprocessor computers, data structures capable of supporting several processes are a growing need. Concurrent data structures seek to provide similar performance to sequential data structures while being accessible concurrently by several processes and providing synchronization mechanisms transparently to those processes. Red-black trees are an important data structure used in many systems. Unfortunately, it is difficult to implement an efficient concurrent red-black tree for shared memory processes; so most research efforts have been directed towards other dictionary data structures, mainly skip-lists and AVL trees. In this paper we present a new type of concurrent red-black tree that uses optimistic concurrency techniques and new balancing operations to scale well and support contention. Our tree performs favorably compared to other similar dictionaries; in particular, in high contention scenarios it performs up to 14% better than the best-known concurrent dictionary solutions.
Embedded hard real time systems require substantial amount of emergency processing power for the management of large scale systems like a nuclear power plant under the threat of an earth quake or a future transport systems under a peril. In order to meet a fully coordinated supervisory control of multiple domains of a large scale system, it requires the scenario of engaging multiprocessor real time design. There are various types of scheduling schemes existing for meeting the critical task assignment in multiple processor environments and it requires the tracking of faulty conditions of the subsystem to avoid system underperformance from failure patterns. Hybrid scheduling usually engages a combined scheduling philosophy comprising of a static scheduling of a set of tasks and a highly pre-emptive scheduling for another set of tasks in different situations of process control. There are instances where highly critical tasks need to be introduced at a least expected catastrophe and it cannot be ensured to meet all deadline in selected processors because of the arrival pattern of such tasks and they bear low tolerance of time to meet the required target. In such circumstances an effective switching of processors for this set of task is feasible and we describe a method to achieve this effectively.
Research in peer-to-peer file sharing systems has focused on tackling the design constraints encountered in distributed systems, while little attention has been devoted to the user experience: these systems always assume the user knows the public key of the file they are searching.
Cloud/Grid computing is envisioned to be a predominant computing model of the future. The movement of files between cloud and client is intrinsic to this model. With the creation of ever expanding data sets, the sizes of files have increased dramatically. Consequently, terabyte file transfers are expected to be the “next big” Internet application. This application is different from other Internet applications in that it requires extensive bandwidth, orders of magnitude larger than the bandwidth requirements of existing applications. It is essential to determine whether or not existing network infrastructures can handle the augmented workload that terabyte transfers would create. This is particularly critical for academic campus networks that are already under strain from high user demand. The paper evaluates the system level challenges of incorporating terabyte transfers into an existing campus network. The evaluation finds that large file transfers can be handled by the current campus network without making major changes to the infrastructure. It is vital to employ a system level service that schedules and monitors the terabyte transfers on users’ behalf. By removing control from users, the service is able to leverage low demand periods and dynamically repurpose unused bandwidth.
This paper studies the general properties for a class of queuing network models that at the thread level, characterize a class of single-core multithreaded processors with various thread scheduling disciplines and parallel resources.
