Enabling HPC for QoS-sensitive Applications: The MANGO Approach

Abstract

Abstract-In this paper, we provide an overview of the MANGO project and its goal. The MANGO project aims at addressing power, performance and predictability (the PPP space) in future High-Performance Computing systems. It starts from the fundamental intuition that effective techniques for all three goals ultimately rely on customization to adapt the computing resources to reach the desired Quality of Service (QoS). From this starting point, MANGO will explore different but interrelated mechanisms at various architectural levels, as well as at the level of the system software. In particular, to explore a new positioning across the PPP space, MANGO will investigate system-wide, holistic, proactive thermal and power management aimed at extreme-scale energy efficiency.

The Italian research on HPC key technologies across EuroHPC

“…Mango. The MANGO FPGA-based manycore emulation platform has been developed in the framework of an H2020 FETHPC project [17]. The platform provides a large-scale cluster of multi-FPGA boards intended for experimenting with customized manycore systems, at the level of both processor and interconnect/system architecture, along with the supporting software stack.…”

Predictive Resource Management for Next-Generation High-Performance Computing Heterogeneous Platforms

“…Once the task-graph is available on the resource manager side, the mapping policy is invoked. After the policy execution, the task-graph encloses the resource mapping information, that will be exploited by the programming library to transparently perform task offloading and buffer allocation, by using the API provided by the underlying HN library (libhn) [7]. For each kernel k ∈ K there is a set of preferred target processing architectures Arch pref s (k) =< Arch 0 , ..., Arch l > that is noted by developer.…”

The RECIPE approach to challenges in deeply heterogeneous high performance systems

“…cloud-based solutions [2], potentially open to a large audience of HPC users. These emerging scenarios require that new HPC platforms are able to support multiple applications running concurrently, possibly with conflicting Quality of Service requirements [3][4][5][6][7]. In addition, at the technology level, RECIPE addresses deep heterogeneity, based on dedicated accelerators like GPUs and FPGAs, as an enabling factor for improved energy efficiency, building on the results collected from previous research projects [8,9].…”