Algorithms for Optimising Heterogeneous Cloud Virtual Machine Clusters

Various types of scalable and high performance computing systems are developed to realize distributed applications. In order to realize not only scalable but also high performance information systems, serverclusterc systems are widely used. In server cluster systems, the larger electric energy is consumed since application processes are performed on the larger number of servers. The improved delay time-based (IDTB)algorithm is discussed to select a server for each request process in a homogeneous server cluster so that the total energy consumption of the homogeneous server cluster to perform application processes can be reduced. In a homogeneous cluster, every server follows the same computation model and power consumption model. In heterogeneous cluster, servers follow different types of computation models and power consumption models. In this paper, we propose an extended IDTB (EIDTB) algorithm to reduce the total energy consumption and response time of each process in heterogeneous server cluster. We evaluate the EIDTB algorithm compared with the IDTB algorithm in a heterogeneous server cluster. In the evaluation, we show the average total energy consumption of a heterogeneous server cluster in the EIDTB algorithm can be reduced maximum 57% of the IDTB algorithm.

In data center companies, cloud computing can host multiple types of heterogeneous virtual machines (VMs) and provide many features, including flexibility, security, support, and even better maintenance than traditional centers. However, some issues need to be considered, such as the optimization of energy usage, utilization of resources, reduction of time consumption, and optimization of virtual machine placement. Therefore, this paper proposes an alternative multiobjective optimization (MOP) approach that combines the salp swarm and sine-cosine algorithms (MOSSASCA) to determine a suitable solution for virtual machine placement (VMP). The objectives of the proposed MOSSASCA are to maximize mean time before a host shutdown (MTBHS), to reduce power consumption, and to minimize service level agreement violations (SLAVs). The proposed method improves the salp swarm and the sine-cosine algorithms using an MOP technique. The SCA works by using a local search approach to improve the performance of traditional SSA by avoiding trapping in a local optimal solution and by increasing convergence speed. To evaluate the quality of MOSSASCA, we perform a series of experiments using different numbers of VMs and physical machines. The results of MOSSASCA are compared with well-known methods, including the nondominated sorting genetic algorithm (NSGA-II), multiobjective particle swarm optimization (MOPSO), a multiobjective evolutionary algorithm with decomposition (MOEAD), and a multiobjective sine-cosine algorithm (MOSCA). The results reveal that MOSSASCA outperforms the compared methods in terms of solving MOP problems and achieving the three objectives. Compared with the other methods, MOSSASCA exhibits a better ability to reduce power consumption and SLAVs while increasing MTBHS. The main differences in terms of power consumption between the MOSCA, MOPSO, MOEAD, and NSGA-II and the MOSSASCA are 0.53, 1.31, 1.36, and 1.44, respectively. Additionally, the MOSSASCA has higher MTBHS value than MOSCA, MOPSO, MOEAD, and NSGA-II by 362.49, 274.70, 585.73 and 672.94, respectively, and the proposed method has lower SLAV values than MOPSO, MOEAD, and NSGA-II by 0.41, 0.28, and 1.27, respectively.

In fault-tolerant information systems, application processes have tobe reliably performed in presence of server faults. Server clustersystems are widely used to realize not only high performance but alsofault-tolerant service. In order to reliably perform applicationprocesses, multiple replicas of each application process are performedon multiple servers in a server cluster system. On the other hand, thelarger amount of electric energy is consumed in a server clustersystem since multiple replicas of each application process areperformed on multiple servers. The redundant delay time based (RDTB) algorithm is discussed to select multiple servers to redundantlyperform each application process in a homogeneous server cluster sothat the total energy consumption of the server cluster can bereduced. In a homogeneous cluster, every server follows the samecomputation model and power consumption model. On the other hand, servers follow different types of computation models and powerconsumption models in a heterogeneous server cluster. Hence, a loadbalancer cannot precisely estimate the processing load and energyconsumption of each server in a heterogeneous server cluster by theRDTB algorithm. In this paper, we propose an improved RDTB (IRDTB) algorithm to select multiple servers for redundantly performing eachapplication process so that the total energy consumption of a servercluster and response time of each process in a heterogeneous servercluster can be reduced. We evaluate the IRDTB algorithm compared withthe RDTB algorithm in a heterogeneous server cluster. In theevaluation, we show the average total energy consumption of aheterogeneous server cluster in the IRDTB algorithm can be reducedmaximum 44% of the RDTB algorithm.

In recent years, increasing attention has been given to the possibility of guaranteeing Service Level Objectives (SLOs) to users about their applications, either regarding performance or power consumption. SLO can be implemented for parallel applications since they can provide many control knobs (e.g., the number of threads to use, the clock frequency of the cores, etc.) to tune the performance and power consumption of the application. Different from most of the existing approaches, we target sequential stream processing applications by proposing a solution based on C++ annotations. The user specifies which parts of the code to parallelize and what type of requirements should be enforced on that part of the code. Our solution first automatically parallelizes the annotated code and then applies self-adaptation approaches at run-time to enforce the user-expressed objectives. We ran experiments on different real-world applications, showing its simplicity and effectiveness.

To attract more customers, public cloud providers offer virtual machine (instance) types that trade off lower prices for poorer capacities. As one salient approach, the providers employ aggressive statistical multiplexing of multiple cheaper instances on a single physical server, resulting in tenants experiencing higher dynamism in the resource capacity of these instances. Examples of this are EC2's "type" instances and GCE's "shared-core" instances.We collectively refer to these as burstable instances for their ability to dynamically "burst" (increase the capacity of) their resources. Burstable instances are significantly cheaper than the "regular" instances, and offer time-varying CPU capacity comprising a minimum guaranteed base capacity/rate, which is much smaller than a short-lived peak capacity that becomes available upon operating at lower than base rate for a sufficient duration. Table 1 summarizes our classification of resource capacity dynamism for GCE and EC2 instances along with the nature of disclosure made by the provider. To exploit burstable instances cost-effectively, a tenant would need to carefully understand the significant additional complexity of such instances beyond that disclosed by the providers.

Heterogeneous clusters that include accelerators have become more common in the realm of high performance computing because of the high GFlop/s rates such clusters are capable of achieving. However, heterogeneous clusters are typically considered hard to program as they usually require programmers to interleave architecture-specific code within application code. We have extended the Charm++ programming model and runtime system to support heterogeneous clusters (with host cores that differ in their architecture) that include accelerators. We are currently focusing on clusters that include commodity processors, Cell processors, and Larrabee devices. When our extensions are used to develop code, the resulting code is portable between various homogeneous and heterogeneous clusters that may or may not include accelerators. Using a simple example molecular dynamics (MD) code, we demonstrate our programming model extensions and runtime system modifications on a heterogeneous cluster comprised of Xeon and Cell processors. Even though there is no architecture-specific code in the example MD program, it is able to successfully make use of three core types, each with a different ISA (Xeon, PPE, SPE), three SIMD instruction extensions (SSE, AltiVec/VMX and the SPE's SIMD instructions), and two memory models (cache hierarchies and scratchpad memories) in a single execution. Our programming model extensions abstract away hardware complexities while our runtime system modifications automatically adjust application data to account for architectural differences between the various cores.

The optimization of resource is crucial for the operation of public cloud systems such as Microsoft Azure, as well as servers dedicated to the workloads of large customers such as Microsoft 365. Those optimization tasks often need to take unknown parameters into consideration and can be formulated as Prediction+Optimization problems. This paper proposes a new Prediction+Optimization method named Correlation-Aware Heuristic Search (CAHS) that is capable of accounting for the uncertainty in unknown parameters and delivering effective solutions to difficult optimization problems. We apply this method to solving the predictive virtual machine (VM) provisioning (PreVMP) problem, where the VM provisioning plans are optimized based on the predicted demands of different VM types, to ensure rapid provisions upon customers' requests and to pursue high resource utilization. Unlike the current state-of-the-art PreVMP approaches that assume independence among the demands for different VM types, CAHS incorporates demand correlation when conducting prediction and optimization in a novel and effective way. Our experiments on two public benchmarks and one industrial benchmark demonstrate that CAHS can achieve better performance than its nine state-of-the-art competitors. CAHS has been successfully deployed in Microsoft Azure and significantly improved its performance. The main ideas of CAHS have also been leveraged to improve the efficiency and the reliability of the cloud services provided by Microsoft 365.

Hundreds of thousands of servers from data centers are operated to provide users with pay-as-yougo infrastructure as a service, platform as a service, and software as a service. Many different types of virtual machine (VM) instances hosted on these servers oftentimes need to efficiently communicate with data movement under current bandwidth capacity. This motivates providers to seek for a bandwidth scheduler to satisfy objectives, namely assuring the minimum bandwidth per VM for the guaranteed deadline and eliminating network congestion as much as possible. Based on some rigorous mathematical models, we formulated a cloud-based bandwidth scheduling algorithm which enables dynamic and fair bandwidth management by categorizing the total bandwidth into several categories and adjusting the allocated bandwidth limit per VM for both upstream and downstream traffics in real time. The simulation showed that paradigm was able to utilize the total assigned bandwidth more efficiently compared to algorithms such as bandwidth efficiency persistence proportional sharing (PPS), PPS, and PS at the network level.

Cloud service providers typically provide different types of virtual machines (VMs) to cloud users with various requirements. Thanks to its effectiveness and fairness, auction has been widely applied in this heterogeneous resource allocation. Recently, several strategy-proof combinatorial cloud auction mechanisms have been proposed. However, they fail to protect the bid privacy of users from being inferred from the auction results. In this paper, we design a differentially private combinatorial cloud auction mechanism (DPCA) to address this privacy issue. Technically, we employ the exponential mechanism to compute a clearing unit price vector with a probability proportional to the corresponding revenue. We further improve the mechanism to reduce the running time while maintaining high revenues, by computing a single clearing unit price, or a subgroup of clearing unit prices at a time, resulting in the improved mechanisms DPCA-S and its generalized version DPCA-M, respectively. We theoretically prove that our mechanisms can guarantee differential privacy, approximate truthfulness and high revenue. Extensive experimental results demonstrate that DPCA can generate near-optimal revenues at the price of relatively high time complexity, while the improved mechanisms achieve a tunable trade-off between auction revenue and running time.

The following topics are dealt with: MIMO communication; array signal processing; radiofrequency interference; probability; wireless channels; iterative methods; OFDM modulation; optimisation; channel estimation; precoding.

The characteristics of cloud computing such as on-demand provisioning of virtual machines in a pay-as-you-go manner have attracted more and more scientific workflows deploying on cloud platforms. Since there are many types of virtual machines which are charged by time intervals, the difficulties of resource provisioning hinders efficient execution of scientific workflows on cloud platforms. To address the challenge, a novel task backfill based scientific workflow scheduling strategy is proposed in this paper. The strategy will use task backfill algorithm to aggregate multiple tasks on a virtual machine instance with suitable performance and fill idle time slot of virtual machines with single tasks, improving resource utilization without affecting the overall performance. Experimental results demonstrate that in comparison with widely used HEFT and IC-PCPD2 strategy, the proposed strategy can effectively reduce the execution cost of the scientific workflows and improve resource utilization while satisfying the deadline constraint.

Business value-aware task scheduling for hybrid IaaS cloud

These days, more and more web are converting into mobile devices such as the internet of things (IoT). These applications are distributed in nature and installed on mobile and servers to achieve many business goals. With this motivation, mobile edge computing (MEC) is an evolving paradigm that reduces the energy consumption of mobile devices through the offload process. However, currently all studies only concentrated on mobile energy and widely overlooked the energy usage of the computing node resources and application deadlines during offloading and scheduling in the MEC network. This research suggests a new mobile edge cloud (mob-cloud) architecture that minimises the nodes energy and meets the deadline requirements of applications. This research formulates this problem as the convex optimisation problem under linear integer programming. The mob-cloud consists of different types of virtual machines which are implementing at the edge computing. This study introduces a novel deadline and energy-efficient task scheduling (DEETS) algorithm framework consisting of task sequencing, resource searching, and mobility-aware scheduling. Furthermore, DEETS keeps tradeoff performances between resource energy usage and task deadline. Simulation results show that the proposed DEETS outperformed all contemporary methods regarding energy usage and deadlines of applications by 50%.

Size and composition of clusters produced by adiabatic expansion of binary gas mixtures (Ar-Kr, Kr-Xe, and N2-Ar) with various component concentrations are studied by using electron-diffraction technique. The resulting homogeneous and heterogeneous clusters are shown to have compositions substantially different from those of the primary gas mixtures and dependent on cluster size. We have found that the key parameters needed for an analysis of cluster composition are the critical cluster radius and the heavier component concentration in the gas mixture that can be used to establish the regions of existence of homogeneous and heterogeneous clusters. These critical values determine the coefficient of the enrichment of clusters with the heavier component with respect to its concentration in the primary gas mixture. Theoretical relations are obtained which provide a good quantitative description of the experimental results and can be expected to be also valid for finding the composition of binary clusters of other van der Waals systems.

In large-scale distributed computing clusters, such as Amazon EC2, there are several types of “system noise” that can result in major degradation of performance: system failures, bottlenecks due to limited communication bandwidth, latency due to straggler nodes, and so on. There have been recent results that demonstrate the impact of coding for efficient utilization of computation and storage redundancy to alleviate the effect of stragglers and communication bottlenecks in homogeneous clusters. In this paper, we focus on general heterogeneous distributed computing clusters consist of a variety of computing machines with different capabilities. We propose a coding framework for speeding up distributed computing in heterogeneous clusters by trading redundancy for reducing the latency of computation. In particular, we propose heterogeneous coded matrix multiplication (HCMM) algorithm for performing distributed matrix multiplication over heterogeneous clusters that are provably asymptotically optimal for a broad class of processing time distributions. Moreover, we show that HCMM is unboundedly faster than any uncoded scheme that partitions the total workload among the workers. To demonstrate how the proposed HCMM scheme can be applied in practice, we provide results from numerical studies and Amazon EC2 experiments comparing HCMM with three benchmark load allocation schemes—uniform uncoded, load-balanced uncoded, and uniform coded. In particular, in our numerical studies, HCMM achieves speedups of up to 73%, 56%, and 42%, respectively, over the three benchmark schemes mentioned earlier. Furthermore, we carry out experiments over Amazon EC2 clusters and demonstrate how HCMM can be combined with rateless codes with nearly linear decoding complexity. In particular, we show that HCMM combined with the Luby transform codes can significantly reduce the overall execution time. HCMM is found to be up to 61%, 46%, and 36% faster than the aforementioned three benchmark schemes, respectively. Additionally, we provide a generalization to the problem of optimal load allocation in heterogeneous settings, where we take into account the monetary costs associated with distributed computing clusters. We argue that HCMM is asymptotically optimal for budget-constrained scenarios as well. In particular, we characterize the minimum possible expected cost associated with a computation task over a given cluster of machines. Furthermore, we develop a heuristic algorithm for (HCMM) load allocation for the distributed implementation of budget-limited computation tasks.