ElCore: Dynamic elastic resource management and discovery for future large-scale manycore enabled distributed systems

Abstract

The next generation of many-core enabled large-scale computing systems relies on thousands of billions of heterogeneous processing cores connected to form a single computing unit. In such large-scale computing environments, resource management is one of the most challenging, and complex issues for efficient resource sharing and utilization, particularly as we move toward Future ManyCore Systems (FMCS). This work proposes a novel resource management scheme for future peta-scale many-core-enabled computing systems, based on hybrid adaptive resource discovery, called ElCore. The proposed architecture contains a set of modules which will dynamically be instantiated on the nodes in the distributed system on demand. Our approach provides flexibility to allocate the required set of resources for various types of processes/applications. It can also be considered as a generic solution (with respect to the general requirements of large scale computing environments) which brings a set of interesting features (such as auto-scaling, multitenancy, multi-dimensional mapping, etc,.) to facilitate its easy adaptation to any distributed technology (such as SOA, Grid and HPC many-core). The achieved evaluation results assured the significant scalability and the high-quality resource mapping of the proposed resource discovery and management over highly heterogeneous, hierarchical and dynamic computing environments with respect to several scalability and efficiency aspects while supporting flexible and complex queries with guaranteed discovery results accuracy. The simulation results prove that, using our approach, the mapping between processes and resources can be done with high level of accuracy which potentially leads to a significant enhancement in the overall system performance.

Resource discovery for distributed computing systems: A comprehensive survey

“…In this section, we discuss these issues, and accordingly, we provide a set of evaluation aspects which must be considered in the evaluation of any discovery approach. Table 7 provides an overview of the most important performance factors (i.e., evaluation metrics) for assessing discovery protocols [295][296][297][298][299]. In the remaining of this section, we further discuss these issues and relevant evaluation strategies to measure the performance factors.…”

HARD: Hybrid Adaptive Resource Discovery for Jungle Computing

“…This provides motivation to go beyond the Grids through the concept of distributed operation systems (DOSs), capable to run on the aforementioned future computing systems. For such DOSs, we used as reference the systems envisaged in the S[o]OS [9][10][11][12][13][14][15][16][17] project, and propose a thread-level resource discovery approach which works in a fully decentralized, self-determining and autonomous fashion, being able to automatically and efficiently alter any P2P-like underlying computing system to a hierarchically structured overly. With respect to decentralization and referring to our target computing environment, described above, it is not guaranteed for a processor to have information about all other processors (or even one processor) in the system.…”

“…Listing 2 presents a description example, using Grammar [5] PT "Processor Type" : bitstring(2,value) {"CPU"=0, "GPU"=1, "FPGA"=2, "Others"=3} , [6] ISA "Instruction Set Architecture" : bitstring(3,value) {" X86"=0, "SPARC"=1, "ARM"=2, "XCORE"=3, "RISC"=4, "CISC"=5, "Legacy"=6, "Others"=7} end layer Layer3 [7] MS "Memory Size by MB" : number, [8] DC "Die Count" : number end attributes [9] INB "Inter-node Bandwidth Mb/S" : number, [10] INL "Inter-node Latency by ns" : number, [11] IGB "Inter-group Bandwidth Mb/S" : number, [12] IGL "Inter-group Latency by ns" : number end end Listing 2: An example of definitions for layers and attributes in .def files…”

“…The authors acknowledge the support of project FP7-ICT-2009.8.1, Grant Agreement No.248465, Service-oriented Operating Systems (2010-2013) [9][10][11][12][13][14][15][16][17] and of project Cloud Thinking (2013-2015), CENTRO-07-ST24-FEDER-002031 [98].…”

“…This work was developed under the framework of the S[o]OS (Service-oriented Operating System) project [9][10][11][12][13][14][15][16][17], European research project aiming to generate a reference addressing architectures for future very large scale distributed infrastructures. The remainder of the paper is structured as follows.…”