We present a multi-GPU design, implementation and performance evaluation of the Halevi-Polyakov-Shoup (HPS) variant of the Fan-Vercauteren (FV) levelled Fully Homomorphic Encryption (FHE) scheme. Our design follows a data parallelism approach and uses partitioning methods to distribute the workload in FV primitives evenly across available GPUs. The design is put to address space and runtime requirements

The scheme [1] is flawed because: (1) its circuit access structure is confusingly described; (2) the cloud server cannot complete the related computations; (3) some users can conspire to generate new decryption keys, without the help of the key generation authority.

Reduce is a fundamental computing pattern, which is widely involved in scientific and engineering applications. For example, accumulation, the most common example of reduce pattern, is the core of applications such as dot product, matrix multiplication, and finite impulse response (FIR) filter. However, there is a trade-off between performance and area in the hardware implementation of the reduce pattern

In this article, a parallel structured divide-and-conquer (PSDC) eigensolver is proposed for symmetric tridiagonal matrices based on ScaLAPACK and a parallel structured matrix multiplication algorithm, called PSMMA. Computing the eigenvectors via matrix-matrix multiplications is the most computationally expensive part of the divide-and-conquer algorithm, and one of the matrices involved in such multiplications

This article presents Egeria, the first automatic synthesizer of advising tools for High-Performance Computing (HPC). When one provides it with some HPC programming guides as inputs, Egeria automatically constructs a text retrieval tool that can advise on what to do to improve the performance of a given program. The advising tool provides a concise list of essential rules automatically extracted from

Control and management in smart grids are facing many challenges such as scalability, heterogeneity and technology innovation. This requires a transformation from the traditional centralised paradigm into a distributed one. In this article, a new distributed programming methodology, called Centralised Programming and Distributed Execution (CPDE), is proposed. CPDE relies on (i) the abstraction of the

This article considers best checkpointing control realizable in real-world systems, whose mean time between failures (MTBFs) often fluctuate. The considered control scheme is based on equating aggregate checkpointing overhead over an activity sequence of interest (θ) and the expected rework amount after a failure recovery for best checkpointing, called “CHORE” (i.e., ch eckpointing o verhead and r

Network Function Virtualization (NFV) has been emerging as an appealing solution that transforms complex network functions from dedicated hardware implementations to software instances running in a virtualized environment. Due to the numerous advantages such as flexibility, efficiency, scalability, short deployment cycles, and service upgrade, NFV has been widely recognized as the next-generation network

Compute node failures are becoming a normal event for many long-running and scalable MPI applications. Keeping within the MPI standards and applying some of the methods developed so far in terms of fault tolerance, we developed a methodology that allows applications to tolerate failures through the creation of semi-coordinated checkpoints within the RADIC architecture. To do this, we developed the

The amount of data transferred over dedicated and non-dedicated network links has been increasing much faster than the increase in the network capacity. On the other hand, the current data transfer solutions fail to guarantee even the promised achievable transfer throughput. In this article, we propose a novel two-phase dynamic throughput optimization model based on mathematical modeling with offline

In the edge computing (EC) environment, edge servers are deployed at base stations to offer highly accessible computing and storage resources to nearby app users. From the app vendor's perspective, caching data on edge servers can ensure low latency in app users’ retrieval of app data. However, an edge server normally owns limited resources due to its limited size. In this article, we investigate the

The proliferation of IoT in various technological realms has resulted in the massive spurt of unsecured data. The use of complex security mechanisms for securing these data is highly restricted owing to the low-power and low-resource nature of most of the IoT devices, especially at the Edge. In this article, we propose to use blockchains for extending security to such IoT implementations. We deploy

Modern micro-service and container-based cloud-native applications have leveraged multi-tenancy as a first class system design concern. The increasing number of co-located services/workloads into server facilities stresses resource availability and system capability in an unconventional and unpredictable manner. To efficiently manage resources in such dynamic environments, run-time observability and

Binarized Neural Networks (BNN), which significantly reduce computational complexity and memory demand, have shown potential in cost- and power-restricted domains, such as IoT and smart edge-devices, where reaching certain accuracy bars is sufficient and real-time is highly desired. In this article, we demonstrate that the highly-condensed BNN model can be shrunk significantly by dynamically pruning

Data deduplication, as a proven technology for effective data reduction in backup and archiving storage systems, is also showing promises in increasing the logical space capacity for storage caches by removing redundant data. However, our in-depth evaluation of the existing deduplication-aware caching algorithms reveals that they only work well when the cached block size is set to 4 KB. Unfortunately

GPU-based computing systems have become a widely accepted solution for the high-performance-computing (HPC) domain. GPUs have shown highly competitive performance-per-watt ratios and can exploit an astonishing level of parallelism. However, exploiting the peak performance of such devices is a challenge, mainly due to the combination of two essential aspects of multi-GPU execution. On one hand, the

The focus of this article is efficient parallelization of the canonical polyadic decomposition algorithm utilizing the alternating least squares method for sparse tensors on distributed-memory architectures. We propose a hypergraph model for general medium-grain partitioning which does not enforce any topological constraint on the partitioning. The proposed model is based on splitting the given tensor

There are great challenges in performing graph coloring on GPU in general. First, the long-tail problem exists in the recursion algorithm because the conflict (i.e., different threads assign the adjacent nodes to the same color) becomes more likely to occur as the number of iterations increases. Second, it is hard to parallelize the sequential spread algorithm because the color allocation depends on

Federated learning (FL) is a distributed deep learning method that enables multiple participants, such as mobile and IoT devices, to contribute a neural network while their private training data remains in local devices. This distributed approach is promising in the mobile systems where have a large corpus of decentralized data and require high privacy. However, unlike the common datasets, the data

This article proposes a parallel algorithm for computing the arithmetic reduction of $n$ numbers as a set of matrix-multiply accumulate (MMA) operations that are executed simultaneously by GPU tensor cores. The analysis, assuming tensors of size $m \times m$ , shows that the proposed algorithm has a parallel running time of $T(n)=5 log_{m^2}{n}$ and a speedup of $S=\frac{4}{5} log_{2}{m^2}$ over a

Large scale simulations are a key pillar of modern research and require ever-increasing computational resources. Different novel manycore architectures have emerged in recent years on the way towards the exascale era. Performance portability is required to prevent repeated non-trivial refactoring of a code for different architectures. We combine Athena++ , an existing magnetohydrodynamics (MHD) CPU

Microservices are widely used for flexible software development. Recently, containers have become the preferred deployment technology for microservices because of fast start-up and low overhead. However, the container layer complicates task scheduling and auto-scaling in clouds. Existing algorithms do not adapt to the two-layer structure composed of virtual machines and containers, and they often ignore

With the rise of network virtualization, the workloads deployed on data center are dramatically changed to support diverse service-oriented applications, which are in general characterized by the time-bounded service response that in turn puts great burden on the data-center networks. Although there have been numerous techniques proposed to optimize the virtual network allocation in data center, the

In the past few years, with the rapid development of CPU-GPU heterogeneous computing, the issue of task scheduling in the heterogeneous cluster has attracted a great deal of attention. This problem becomes more challenging with the need for efficient co-execution of tasks on the GPUs. However, the uncertainty of heterogeneous cluster and the interference caused by resource contention among co-executing

Edge computing, as an extension of cloud computing, distributes computing and storage resources from centralized cloud to distributed edge servers, to power a variety of applications demanding low latency, e.g., IoT services, virtual reality, real-time navigation, etc. From an app vendor's perspective, app data needs to be transferred from the cloud to specific edge servers in an area to serve the

Data reliability and availability, and serviceability (RAS) of erasure-coded data centers are highly affected by data repair induced by node failures. In a traditional failure identification scheme, all chunks share the same identification time threshold, thus losing opportunities to further improve the RAS. To solve this problem, we propose RAFI, a novel risk-aware failure identification scheme. In

Communication traces collected from MPI applications are an important source of information for performance optimization as they can help analysts determine communication patterns and identify inefficiencies. However, their collection, especially at scale, is time consuming, since it usually requires running the complete target application on a large number of nodes. In this work, we present PredCom

Redundancy schemes are introduced to in-memory stores to provide fault tolerance. To achieve good trade-off between access performance and memory efficiency, it is appropriate to adopt replication and erasure coding to keep popular and unpopular data, respectively. Within such a hybrid-redundancy in-memory store, an issue of redundancy transition from replication to erasure coding (a.k.a., erasure-coded

Large-scale molecular dynamics (MD) simulations on supercomputers play an increasingly important role in many research areas. With the capability of simulating charge equilibration (QEq), bonds and so on, Reactive force field (ReaxFF) enables the precise simulation of chemical reactions. Compared to the first principle molecular dynamics (FPMD), ReaxFF has far lower requirements on computational resources

Prevailing data-intensive applications, such as artificial intelligence and internet of things, demand considerable compute capability. Coarse-grained reconfigurable architectures (CGRAs) can meet this demand via providing abundant compute resources. However, compilation has become an essential problem because the increasing resources need to be orchestrated efficiently. Static compilation is insufficient

NVMe SSDs provide extremely high performance and have been widely deployed in distributed object storage systems in data centers. However, we observe that there are still severe performance degradation and write amplification under the unaligned writes scenario with high-performance SSDs. In this article, we identify that the RMW sequence which is used to handle the unaligned writes incurs severe overhead

The advent of online video streaming applications and services along with the users’ demand for high-quality contents require High Efficiency Video Coding (HEVC), which provides higher video quality and more compression at the cost of increased complexity. On one hand, HEVC exposes a set of dynamically tunable parameters to provide trade-offs among Quality-of-Service (QoS), performance, and power consumption

Chip manufacturers introduce redundancy at various levels of CPU design to guarantee correct operation, even for worst-case combinations of non-idealities in process variation and system operating conditions. This redundancy is implemented partly in the form of voltage margins. However, for a wide range of real-world execution scenarios these margins are excessive and merely translate to increased

Contemporary graphics processing units (GPUs) support dynamic voltage and frequency scaling to balance computational performance and energy consumption. However, accurate and straightforward performance estimation for a given GPU kernel under different frequency settings is still lacking for real hardware, which is essential to determine the best frequency configuration for energy saving. In this article

With the increase of the number of electric vehicles (EVs), it is of vital importance to develop the efficient and effective charging scheduling schemes for all the EVs. In this article, we aim to maximize the social welfare of all the EVs, charging stations (CSs) and power plant (PP), by taking into account the changing demand of each EV, the changing price, the capacity and the congestion balance

Depth estimation from stereo images is essential to many applications such as robotics and autonomous vehicles, most of which ask for the real-time response, high energy and storage efficiency. Recent work has shown deep neural networks (DNN) perform extremely well for stereo estimation. However, these state-of-the-art DNN based algorithms are challenging to be deployed into real-world applications

Understanding the I/O characteristics of supercomputers is crucial for grasping accurate I/O workloads and uncovering potential I/O inefficiency. We collect and analyze I/O traces from two production supercomputers, and find that the I/O traffic peaks in the system not only occur in short periods of time but also originate from a minority of adjacent compute nodes, which we call spatially bursty I/O

The focus is distributed-memory parallelization of sparse-general-matrix-multiplication (SpGEMM). Parallel SpGEMM algorithms are classified under one-dimensional (1D), 2D, and 3D categories denoting the number of dimensions by which the 3D sparse workcube representing the iteration space of SpGEMM is partitioned. Recently proposed successful 2D- and 3D-parallel SpGEMM algorithms benefit from upper

We study an interval job scheduling problem in distributed systems. We are given a set of interval jobs, with each job specified by a size, an arrival time and a processing length. Once a job arrives, it must be placed on a machine immediately and run for a period of its processing length without interruption. The homogeneous machines to run jobs have the same capacity limits such that at any time

We introduce XtraPuLP , a distributed-memory graph partitioner designed to process irregular trillion-edge graphs. XtraPuLP is based on the scalable label propagation community detection technique, which has been demonstrated in various prior works as a viable means to produce high quality partitions of skewed and small-world graphs with minimal computation time. Our XtraPuLP implementation can also

Distributed deep learning systems (DDLS) train deep neural network models by utilizing the distributed resources of a cluster. Developers of DDLS are required to make many decisions to process their particular workloads in their chosen environment efficiently. The advent of GPU-based deep learning, the ever-increasing size of datasets, and deep neural network models, in combination with the bandwidth

We investigate a workflow scheduling problem with stochastic task arrival times and fuzzy task processing times and due dates. The problem is common in many real-time and workflow-based applications, where tasks with fixed stage number and linearly dependency are executed on scalable cloud resources with multiple price options. The challenges lie in proposing effective, stable, and robust algorithms

Data analytics has become an integral part of large-scale scientific computing. Among various data analytics frameworks, MapReduce has gained the most traction. Although some efforts have been made to enable efficient MapReduce for supercomputing systems, they are often limited to fairly homogeneous workloads where equal partitioning of input data across tasks results in essentially equal output or

Expectation maximization (EM) is a clustering-based machine learning algorithm that is widely used in many areas of science (e.g., bioinformatics and computer vision) to find maximum likelihood and maximum a posteriori estimates for models with latent variables. To deploy such an algorithm in cloud environments, security and privacy issues need be considered to avoid data breaches or abuses by external

Two-phase I/O is a well-known strategy for implementing collective MPI-IO functions. It redistributes I/O requests among the calling processes into a form that minimizes the file access costs. As modern parallel computers continue to grow into the exascale era, the communication cost of such request redistribution can quickly overwhelm collective I/O performance. This effect has been observed from

Hardware accelerators have been a promising solution to reduce the cost of cloud datacenters. This article investigates the acceleration of an important datacenter workload: the web server (or proxy) that faces high computational consumption originated from SSL/TLS processing and HTTP compression. Our study reveals that for the widely-deployed event-driven web architecture, the straight offloading

The growing popularity of in-memory computing for bigdata analytics often causes performance bottlenecks to memory subsystem resided in operating systems (OS). This article purposes cooperative memory expansion (COMEX), an OS kernel extension. COMEX establishes a stable pool of memory collectively across nodes in a cluster and enhances OS's memory subsystem for memory aggregation from connected machines

In 2014, Wu et al. proposed a tag encoding scheme, named KEPTE, to protect network coding against pollution attack. They also carefully analyzed the security of KEPTE based on the transmission of a data file through their key-pre-distributed network. In this article, we point out that their security analysis only holds for single data file transmitted in this network. If multiple files are multicasted

Cloud storage security has gained considerable research efforts with the wide adoption of cloud computing. As a security mechanism, researchers have been investigating cloud storage auditing schemes that enable a user to verify whether the cloud keeps the user's outsourced data undamaged. However, existing schemes have usability issues in compatibility with existing real world cloud storage applications

This article provides a comprehensive study of the impact of performance optimizations on the energy efficiency of a real-world CFD application called MPDATA, as well as an insightful analysis of performance-energy interaction of these optimizations with the underlying hardware that represents the first generation of Intel Xeon Scalable processors. Considering the MPDATA iterative application as a

We study job assignment in large, heterogeneous resource-sharing clusters of servers with finite buffers. This load balancing problem arises naturally in today's communication and big data systems, such as Amazon Web Services, Network Service Function Chains, and Stream Processing. Arriving jobs are dispatched to a server, following a load balancing policy that optimizes a performance criterion such

The current standalone deep learning framework tends to result in overfitting and low utility. This problem can be addressed by either a centralized framework that deploys a central server to train a global model on the joint data from all parties, or a distributed framework that leverages a parameter server to aggregate local model updates. Server-based solutions are prone to the problem of a sin

The Last Level Cache (LLC) plays a key role in the system performance of current multi-cores by reducing the number of long latency main memory accesses. The inter-application interference at this shared resource, however, can lead the system to undesired situations regarding performance and fairness. Recent approaches have successfully addressed fairness and turnaround time (TT) in commercial processors

Delivering optimum performance on a parallel computer is highly dependant on the efficiency of the scheduling and mapping procedure. If the composition of the parallel application is known a prior, the mapping can be accomplished statically on the compilation time. The mapping algorithm uses the model of the parallel application and maps its tasks to processors in a way to minimize the total execution

SPHINCS was recently proposed as a stateless, quantum-resilient hash-based signature scheme. However, one possible limitation of SPHINCS is its signing speed, namely, the best known implementation merely produces a few hundred of signatures per second, which is not good enough, e.g., for a social website with a huge amount of users. Aiming at improving the singing throughput, we present highly parallel

In the distributed computing framework of Spark, cross-node/rack data transfer produced by map tasks and reduce tasks are common problems resulting in performance degradation, such as prolonging of entire execution time and network congestion. To address these problems, this article utilizes the bipartite graph modelling to propose an optimal locality-aware task scheduling algorithm. By considering

Solid State Drives (SSDs) are popularly used for caching in large scale cloud storage systems nowadays. Traditionally, most cache algorithms make replacement upon each miss when cache space is full. However, we observe that in a typical Cloud Block Storage (CBS) system, there is a great percentage of blocks with large reuse distances, which would result in large number of blocks being evicted out of

Spiking neural network (SNN) is the most commonly used computational model for neuroscience and neuromorphic computing communities. It provides more biological reality and possesses the potential to achieve high computational power and energy efficiency. Because existing SNN simulation frameworks on general-purpose graphics processing units (GPGPUs) do not fully consider the biological oriented properties

The proliferation of system virtualization poses a new challenge for the coarse-grained time sharing techniques for consolidation, since operating systems are running on virtual CPUs. The current system stack was designed under the assumption that operating systems can seize CPU resources at any moment. However, for the guest operating system on a virtual machine (VM), such assumption cannot be guaranteed

We introduce AL4SAN , a lightweight library for abstracting the APIs of task-based runtime engines. AL4SAN unifies the expression of tasks and their data dependencies. It supports various dynamic runtime systems relying on compiler technology and user-defined APIs. It enables a single application to employ different runtimes and their respective scheduling components, while providing user-obliviousness