Interval Type-2 Fuzzy Sets (IT2 FSs) has been used in wide range of applications, such as decision making, pattern recognitions. However, there is little investigation on the clustering techniques of IT2 FSs, since it...
详细信息
Interval Type-2 Fuzzy Sets (IT2 FSs) has been used in wide range of applications, such as decision making, pattern recognitions. However, there is little investigation on the clustering techniques of IT2 FSs, since it is very important for measuring the degree of similarity between interval type-2 fuzzy sets in clustering research. In this paper, we give a novel method to represent interval type-2 fuzzy sets by a series of linear functions. This method is named as triangular representation because the curves of the functions and the coordinates shape into triangles. A new kind of similarity between interval type-2 fuzzy sets based on triangle is given and the validity of this method is proven. Furthermore, several numerical examples are given to show that the proposed similarity measure is reasonable and better than any existing methods.
As the degree of virtualization is growing considerably, improving performance of virtual machine environments motivates deeper investigation of the internal processes and performance implications of virtualization. S...
详细信息
As the degree of virtualization is growing considerably, improving performance of virtual machine environments motivates deeper investigation of the internal processes and performance implications of virtualization. Several tools are currently available to help analyze performance in virtual machine environments. However, all of these cannot support users to add their own trace information. In this paper, we present Xenrelay, a unified, efficient, and simple mechanism for transferring large amounts of data from guest domain kernel to privileged domain user-space. Xenrelay allows users (who trace subsystems of guest domain to analyze performance) to record and fast relay data between domains. We provide a case of Xenrelay's value in revealing behavior of virtualization. We build a block trace toolkit (using Xenrelay as an engine). This toolkit is used to track the block I/O virtualization procedures and evaluates overheads in the Xen environment. Our trace results identify that the main delay is the time of request transmission and block device operation. Besides, the results reveal impact of the I/O schedules on the disk performance.
The advent of Persistent Memory (PM) necessitates an evolution of Remote Direct Memory Access (RDMA) technologies for supporting remote data persistence. Previous software-based solutions require remote CPU interventi...
详细信息
ISBN:
(数字)9781450384421
ISBN:
(纸本)9781665483902
The advent of Persistent Memory (PM) necessitates an evolution of Remote Direct Memory Access (RDMA) technologies for supporting remote data persistence. Previous software-based solutions require remote CPU intervention and postpone the visibility of remote persistence. In this paper, we design several hardware-supported RDMA primitives to flush data from the volatile cache of RDMA Network Interface Cards (RNICs) to the PM. We also propose durable RPCs based on the proposed RDMA Flush primitives to support remote data persistence and fast failure recovery. We emulate the performance of RDMA Flush primitives through other RDMA prim-itives, and compare our proposals with several state-of-the-art RPCs in a real testbed equipped with PM and InfiniBand networks. Experimental results show that our proposals can improve the throughput of RPCs by up to 90%, and reduce the 99th percentile latency by up to 49%. The experimental studies also provide instructive guidelines for designing RDMA-based distributed PM systems.
To process large-scale real-time data streams, existing distributed stream processing systems (DSPSs) leverage different stream partitioning strategies. The one-to-many data partitioning strategy plays an important ro...
详细信息
ISBN:
(数字)9781450384421
ISBN:
(纸本)9781665483902
To process large-scale real-time data streams, existing distributed stream processing systems (DSPSs) leverage different stream partitioning strategies. The one-to-many data partitioning strategy plays an important role in various applications. With one-to-many data partitioning, an upstream processing instance sends a generated tuple to a potentially large number of downstream processing instances. Existing DSPSs leverage an instance-oriented communication mechanism, where an upstream instance transmits a tuple to different downstream instances separately. However, in one-to-many data partitioning, multiple downstream instances typically run on the same machine to exploit multi-core resources. Therefore, a DSPS actually sends a data item to a machine multiple times, raising significant unnecessary costs for serialization and communication. We show that such a mechanism can lead to serious performance bottleneck due to CPU overload. To address the problem, we design and implement Whale, an efficient RDMA (Remote Direct Memory Access) assisted distributed stream processing system. Two factors contribute to the efficiency of this design. First, we propose a novel RDMA-assisted stream multicast scheme with a self-adjusting non-blocking tree structure to alleviate the CPU workloads of an upstream instance during one-to-many data partitioning. Second, we re-design the communication mechanism in existing DSPSs by replacing the instance-oriented communication with a new worker-oriented communication scheme, which saves significant costs for redundant serialization and communication. We implement Whale on top of Apache Storm and conduct comprehensive experiments to evaluate its performance with large-scale real world datasets. The results show that Whale achieves 56.6x improvement of system throughput and 97% reduction of processing latency compared to existing designs.
Emerging data centers may host a large number of applications that consume CPU power, memory, and I/O resources. Previous studies focus on the allocation of resources in order to perfectly satisfy the demands seen in ...
详细信息
With the development of cloud computing, there is a growing number of virtual machines (VMs) in the IaaS cloud. The VM owners can install different kinds of software on demand. However, if the software is not updated ...
详细信息
With the development of cloud computing, there is a growing number of virtual machines (VMs) in the IaaS cloud. The VM owners can install different kinds of software on demand. However, if the software is not updated in time, it would be a great threat to the security of the cloud. But for the VM owners, it is a tedious task to keep all of the installed software up to date. In this paper we present a new software update model called UaaS (Update as a Service) to handle the VM (online or offline) update automatically. Multiple VMs share one software update service and multiple update strategies are provided for single software, which can be customized at any time. The ability of UaaS has been evaluated by our experiments, and the results show that UaaS can provide software update service successfully and complete update task for lots of VMs with multiple update strategies efficiently.
Large-scale distributed deep learning is of great importance in various applications. For distributed training, the inter-node gradient communication often becomes the performance bottleneck. Gradient sparsification h...
详细信息
With the development of virtualization technology, it¿s desirable to deploy virtual machines to high performance clusters used for data centers. VNIX, developed in servicescomputingtechnology and systemlab, tr...
详细信息
With the development of virtualization technology, it¿s desirable to deploy virtual machines to high performance clusters used for data centers. VNIX, developed in servicescomputingtechnology and systemlab, tries to help cluster administrators to manage a large number of virtual machines (VMs) distributed on clusters. To reduce the complexity of virtualization management, VNIX provides a whole-set of tools for monitoring, deploying, controlling, and configuring virtual machines on clusters. In addition to those basic management functions, VNIX also offers a number of specialized tools for clustering VMs. Due to the complex dynamic environment in clusters, it¿s challenging to design such tools. In this paper, we present the design of VNIX, and we describe several use cases of managing VMs in clusters with VNIX. Such use cases illustrate various ways of using VNIX to simplify the management work and to improve resource utilization.
暂无评论