Technology (IJRCIT), Vol. 4, Issue 3, June-2019 ISSN: 2455-3743
“ROLE OF PARALLEL PROCESSOR IN CLUSTER COMPUTING”
1SHRUTI BAKSHI
DCPE, P.G Department of Computer Science &Technology, H.V.P.M, Amravati, India [email protected]
2PROF. V. S. BELSARE
DCPE, P.G Department of Computer Science &Technology, H.V.P.M, Amravati, India [email protected]
ABSTRACT: Parallel computing is critical in many areas of computing, from solving complex scientific problems to improving the computing experience for smart device and personal computer users. Clusters of computers have become more popular in recent years as the power of commodity processors has approached that of workstation processors. Yet at the same time, the increased usage of computer technologies has meant that costs of processors as well as other computer sub-systems (such as motherboard, memory, hard disks, network cards) has benefited from the scale-of-economy. The standardization of network hardware and software protocols has also improved the viability of cluster computing. This paper will provide a review of the cluster computing and parallel processing.
Keywords: Parallel computing, cluster computing, workstation processors, smart device
1. INTRODUCTION
Parallel computing is pushing the limits of advancement in registering rate and capacity. Customarily, PCs have run sequential calculations. A PC would execute one guidance adhered to by another guidance, advancing through a lot of directions each one in turn. In parallel figuring, directions can run simultaneously on various processors. Different processors each handle one guidance. In the present individualized computing condition, which incorporates advanced cells, tablets, workstations and PCs, the focal handling unit can have multi-center processors. Parallel figuring can utilize multi- center processors or multi-processor PCs. A computer cluster consists of a set of loosely connected or tightly connected computers that work together so that in many respects they can be viewed as a single system. The components of a cluster are usually connected to each other through fast local area networks with each node running its own instance of an operating system. Computer clusters emerged as a result of convergence of a number of computing trends including the availability of low cost microprocessors, high speed networks, and software for high performance distributed computing.
Clusters are usually deployed to improve performance and availability over that of a single computer, while typically being much more cost-effective than single computers of comparable speed or availability.
2. OVERVIEW OF PARALLEL COMPUTING In cluster system architecture, gatherings of processors are sorted out into hundreds or thousands of hubs, inside which the CPUs convey by means of shared memory. Hubs are interconnected with a correspondence texture that is sorted out as a system. Parallel projects use gatherings of CPUs on at least one hub.
To misuse the intensity of bunch PCs, parallel projects should guide various processors to fathom various pieces of a
calculation all the while. To be effective, a parallel program must be intended for a particular framework design. It additionally should be customized to keep running on frameworks that contrast in the quantity of CPUs associated by shared memory, the quantity of memory store levels, how those reserves are circulated among CPUs, and the qualities of the correspondence component for message passing.
You additionally need to see how to utilize every PC's framework programming and the administrations that help you run your code on that stage. Your capacity to work profitably on these perplexing figuring stages is extraordinarily improved by framework explicit administrations, for example, compilers that offer different degrees of code enhancement, bunch work schedulers, framework asset directors for parallel employments, and upgraded libraries. To become a proficient user of these powerful computing systems, you need to understand all three of these aspects of parallel computing and how they relate to each other:
• System hardware architecture
• User code tailored to computer hardware
• System software and services that enable users to modify their code for various platforms
There are numerous motivations to turn into a capable client of these frameworks and utilize their assets. One is that you utilize the PC time that you get with your allotment. Also, when your code runs effectively, you decrease your activity turnaround time (divider clock time) and your activity has less introduction to execution delays on these vigorously bought in PCs. Lessening your activity's divider clock time makes a greater amount of your time accessible for different exercises, and better utilization of the framework additionally benefits the whole client network.
Technology (IJRCIT), Vol. 4, Issue 3, June-2019 ISSN: 2455-3743
3. PARALLEL PROGRAMMING PARADIGMS Parallel programming paradigms involve two issues:
• Efficient use of CPUs on one process
• Communication between nodes to support interdependent parallel processes running on different nodes and exchanging mutually dependent data A parallel program typically comprises of a lot of procedures that offer information with one another by imparting through shared memory over a system interconnect texture.
Parallel projects that immediate different CPUs to speak with one another by means of shared memory commonly utilize the OpenMP interface. The free activities running on different CPUs inside a hub are called strings.
Parallel projects that immediate CPUs on various hubs to share information must utilize message disregarding the system. These projects utilize the Message Passing Interface (MPI).
At long last, programs that utilization deliberately coded crossover procedures can be equipped for both superior and high effectiveness. These half breed projects utilize both OpenMP and MPI.
4. CLUSTERING FOR MASSIVE PARALLELISM
• A computer cluster is a collection of interconnected stand-alone computers which can work together collectively and cooperatively as a single integrated computing resource pool.
• Clustering explores massive parallelism at the job level and achieves high availability (HA) through stand-alone operations.
• The benefits of computer clusters and massively parallel processors (MPPs) include scalable performance, HA, fault tolerance, modular growth, and use of commodity components.
• These features can sustain the generation changes experienced in hardware, software, and network components.
• Cluster computing became popular in the mid-1990s as traditional mainframes and vector supercomputers were proven to be less cost-effective in many high- performance computing (HPC) applications.
5. DESIGN OBJECTIVES OF CLUSTERS COMPUTING
Clusters have been classified in various ways in the literature.
We classify clusters using six orthogonal attributes:
scalability, packaging, control, homogeneity, programmability, and security.
Scalability
• Clustering of computers is based on the concept of modular growth.
• To scale a cluster from hundreds of uni-processor nodes to a super cluster with 10,000 multi-core nodes is a nontrivial task.
• The scalability could be limited by a number of factors, such as the multi-core chip technology, cluster topology, packaging method, power consumption, and cooling scheme applied.
• The purpose is to achieve scalable performance constrained by the aforementioned factors.
• We have to also consider other limiting factors such as the memory wall, disk I/O bottlenecks, and latency tolerance, among others.
Packaging
• Cluster nodes can be packaged in a compact or a slack fashion.
• In a compact cluster, the nodes are closely packaged in one or more racks sitting in a room, and the nodes are not attached to peripherals (monitors, keyboards, mice, etc.).
• In a slack cluster, the nodes are attached to their usual peripherals (i.e., they are complete SMPs, workstations, and PCs), and they may be located in different rooms, different buildings, or even remote regions.
• Packaging directly affects communication wire length, and thus the selection of interconnection technology used.
• While a compact cluster can utilize a high- bandwidth, low-latency communication network that is often proprietary, nodes of a slack cluster are normally connected through standard LANs or WANs.
Control
• A cluster can be either controlled or managed in a centralized or decentralized fashion.
• A compact cluster normally has centralized control, while a slack cluster can be controlled either way.
• In a centralized cluster, all the nodes are owned, controlled, managed, and administered by a central operator.
• In a decentralized cluster, the nodes have individual owners. For instance, consider a cluster comprising an interconnected set of desktop workstations in a department, where each workstation is individually owned by an employee.
• The owner can reconfigure, upgrade, or even shut down the workstation at any time. This lack of a single point of control makes system administration of such a cluster very difficult.
Technology (IJRCIT), Vol. 4, Issue 3, June-2019 ISSN: 2455-3743
• It also calls for special techniques for process scheduling, workload migration, check pointing, accounting, and other similar tasks.
Homogeneity
• A homogeneous cluster uses nodes from the same platform, that is, the same processor architecture and the same operating system; often, the nodes are from the same vendors.
• A heterogeneous cluster uses nodes of different platforms. Interoperability is an important issue in heterogeneous clusters.
• For instance, process migration is often needed for load balancing or availability. In a homogeneous cluster, a binary process image can migrate to another node and continue execution.
• This is not feasible in a heterogeneous cluster, as the binary code will not be executable when the process migrates to a node of a different platform.
Security
• Intra-cluster communication can be either exposed or enclosed.
• In an exposed cluster, the communication paths among the nodes are exposed to the outside world.
An outside machine can access the communication paths, and thus individual nodes, using standard protocols (e.g., TCP/IP).
• Such exposed clusters are easy to implement, but have several disadvantages:
• Being exposed, intra cluster communication is not secure, unless the communication subsystem performs additional work to ensure privacy and security. Outside communications may disrupt intra cluster communications in an unpredictable fashion.
• In an enclosed cluster, intra cluster communication is shielded from the outside world, which alleviates the aforementioned problems. A disadvantage is that there is currently no standard for efficient, enclosed intra cluster communication.
6. ARCHITECTURE OF CLUSTER COMPUTING A cluster computer is a type of parallel or distributed processing system, which consists of a collection of interconnected stand-alone computers working together as a single integrated resource. The typical architecture of a cluster is shown figure 1.
Figure 1: Architecture of Cluster Computer 7. CLUSTER DESIGN ISSUES
Scalable Performance: This refers to the fact that scaling of resources (cluster nodes, memory capacity, I/O bandwidth, etc.) leads to a proportional increase in performance. Of course, both scale-up and scale down capabilities is needed, depending on application demand or cost effectiveness considerations. Clustering is driven by scalability.
Single-System Image (SSI): A set of workstations connected by an Ethernet network is not necessarily a cluster. A cluster is a single system. For example, suppose a workstation has a 300 Mflops/second processor, 512 MB of memory, and a 4 GB disk and can support 50 active users and 1,000 processes. By clustering 100 such workstations, can we get a single system that is equivalent to one huge workstation, or a mega-station, that has a 30 Gflops/second processor, 50 GB of memory, and a 400 GB disk and can support 5,000 active users and 100,000 processes? SSI techniques are aimed at achieving this goal.
Availability Support: Clusters can provide cost-effective HA capability with lots of redundancy in processors, memory, disks, I/O devices, networks, and operating system images.
However, to realize this potential, availability techniques are required.
Cluster Job Management: Clusters try to achieve high system utilization from traditional workstations or PC nodes that are normally not highly utilized. Job management software is required to provide batching, load balancing, parallel processing, and other functionality. Special software tools are needed to manage multiple jobs simultaneously.
8. TYPES OF PARALLELISM
Bit-level parallelism: It is the form of parallel computing which is based on the increasing processor’s size. It reduces the number of instructions that the system must execute in order to perform a task on large-sized data. Example: Consider a scenario where an 8-bit processor must compute the sum of
Technology (IJRCIT), Vol. 4, Issue 3, June-2019 ISSN: 2455-3743
two 16-bit integers. It must first sum up the 8 lower-order bits, then add the 8 higher-order bits, thus requiring two instructions to perform the operation. A 16-bit processor can perform the operation with just one instruction.
Instruction level parallelism: A processor can only address less than one instruction for each clock cycle phase. These instructions can be re-ordered and grouped which are later on executed concurrently without affecting the result of the program. This is called instruction-level parallelism.
Task Parallelism: Task parallelism employs the decomposition of a task into subtasks and then allocating each of the subtasks for execution. The processors perform execution of sub tasks concurrently.
9. CLUSTER PARALLEL PROCESSING OFFERS SEVERAL IMPORTANT ADVANTAGES
• Each of the machines in a group can be a finished framework, usable for a wide scope of other figuring applications. This leads numerous individuals to recommend that bunch parallel registering can essentially guarantee all the "squandered cycles" of workstations sitting inert on individuals' work areas.
It isn't generally so natural to rescue those cycles, and it will likely moderate your colleague's screen saver, however it very well may be finished.
• The current blast in organized frameworks implies that the vast majority of the equipment for structure a bunch is being sold in high volume, with correspondingly low "ware" costs as the outcome.
Further investment funds originate from the way that just a single video card, screen, and console are required for each group (in spite of the fact that you may need to swap these into each machine to play out the underlying establishment of Linux, when running, a commonplace Linux PC does not require a
"comfort"). In correlation, SMP and appended processors are a lot littler markets, inclining toward to some degree more expensive rate per unit execution.
• Cluster figuring can scale to huge frameworks. While it is presently elusive a Linux-perfect SMP with a lot multiple processors, most ordinarily accessible system equipment effectively develops a group with to 16 machines. With a little work, hundreds or even a large number of machines can be arranged. Indeed, the whole Internet can be seen as one genuinely tremendous group.
• The actuality that supplanting a "terrible machine"
inside a bunch is paltry contrasted with fixing an incompletely flawed SMP yields a lot higher accessibility for painstakingly planned group designs.
This winds up significant not just for specific applications that can't endure huge administration interferences, yet in addition for general utilization of
frameworks containing enough processors so single- machine disappointments are genuinely normal.
10. CONCLUSION
Cluster computer are not a panacea for the majority of society's registering needs, there exists a wide scope of issues for which group are the perfect arrangement. Since cluster registering would now be able to help a more extensive scope of utilizations, it's turned out to be progressively valuable.
Since cluster would now be able to contain a scope of process hubs, including on-premises servers, work area workstations, and cloud cases, bunch registering has turned out to be increasingly available. For anyone who thought this way to deal with making programming run quick wasn't important to them, it's a great opportunity to look again.
11. REFERENCES
[1] Lai, Chenggang & Hao, Zhijun & Huang, Miaoqing & Shi, Xuan & You, Haihang. (2014). Comparison of Parallel Programming Models on Intel MIC Computer Cluster. 925- 932. 10.1109/IPDPSW.2014.105.
[2] Chee Shin Yeo, Rajkumar Buyya, Hossein Pourreza, Rasit Eskicioglu, Peter Graham, Frank Sommers, "Cluster Computing: High-Performance, High-Availability, and High- Throughput Processing on a Network of Computers", 2006/01/01, 10.1007/0-387-27705-6_16.
[3] Leigh Anne Clevenger, Hugh Eng, Kevin Khan, Javid Maghsoudi, and Mantie Reid, "Parallel Computing Hardware and Software Architectures for High Performance Computing", Proceedings of Student-Faculty Research Day, CSIS, Pace University, May 1st, 2015.
[4] S. S. Baghsorkhi, M. Delahaye, S. J. Patel, W. D. Gropp, and W. W. Hwu. An adaptive performance modeling tool for GPU architectures. In Proc. of the 15th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, 2010.
[5] B. Barney, "Introduction to Parallel Computing."
Introduction to Parallel Computing. Lawrence Livermore National Laboratory, 14 July 2014. Web. 24 Sept. 2014.
<https://computing.llnl.gov/tutorials/paral-lel_comp/>.
[6] E. Betti, M. Cesati, R. Gioiosa, and F. Piermaria, “A global operating system for HPC clusters,” in IEEE International Conference on Cluster Computing and Workshops, 2009.
CLUSTER ’09, 2009, pp. 1–10.
[7] H. Fang, Z. Yue-Long. File Routing in an Intelligent Network Disk. 2012 International Workshop on Education Technology and Computer Science; pp. 485-87
[8] M. K. Gobbert, “Configuration and performance of a Beowulf cluster for large-scale scientific simulations,”
Technology (IJRCIT), Vol. 4, Issue 3, June-2019 ISSN: 2455-3743
Computing in Science Engineering, vol. 7, no. 2, pp. 14–26, Mar. 2005.
[9] Intel, “THE MANYCORE SHIFT: Microsoft Parallel Computing Initiative Ushers Computing into the Next Era”, http://www.intel.com/press-room/kits/upcrc/
ParallelComputing_backgrounder.pdf, 2014, accessed November 2014
