The proliferation of cloud computing has promoted the wide deployment of largescale datacenters with tremendous power consumption and high carbon emission. To reduce power cost and carbon footprint, an increasing number of cloud service providers have considered green datacenters with renewable energy sources, such as solar or wind. However, unlike the stable supply of grid energy, it is challenging to utilize and realize renewable energy due to the uncertain, intermittent and variable nature. In this article, we provide a taxonomy of the state-of-the-art research in applying renewable energy in cloud computing datacenters from five key aspects, including generation models and prediction methods of renewable energy, capacity planning of green datacenters, intra-datacenter workload scheduling and load balancing across geographically distributed datacenters. By exploring new research challenges involved in managing the use of renewable energy in datacenters, this article attempts to address why, when, where and how to leverage renewable energy in datacenters, also with a focus on future research avenues.
Datacenters are facing increasing pressure to cap their carbon footprints at low cost. Recent work has shown the significant environmental benefits of using renewable energy for datacenters by supply-following techniques (workload scheduling, geographical load balancing, etc.) However, all such prior work has only considered on-site renewable generation when numerous other options also exist, which may be superior to on-site renewables for many datacenters. Alternative ways for datacenters to incorporate renewable energy into their overall energy portfolio include: construction of or investment into off-site renewable farms at locations with more abundant renewable energy potential, indirect purchase of renewable energy through buying renewable energy certificates (RECs), purchase of renewable energy products such as power purchase agreements (PPAs) or through third-party renewable providers. We propose a general, optimization-based framework to minimize datacenter costs in the presence of different carbon footprint reduction goals, renewable energy characteristics, policies, utility tariff, and energy storage devices (ESDs). We expect that our work can help datacenter operators make informed decisions about sustainable, renewable-energy-powered IT system design.
In this paper, we propose an unorthodox topology for datacenters that eliminates all hierarchical switches in favor of connecting nodes at random according to a small-world-inspired distribution. Specifically, we examine topologies where the underlying nodes are connected at the small scale in a regular pattern, such as a ring, torus or cube, such that every node can route efficiently to nodes in its immediate vicinity, and amended by the addition of random links to nodes throughout the datacenter, such that a greedy algorithm can route packets to far away locations efficiently. Coupled with geographical address assignment, the resulting network can provide content routing in addition to traditional routing, and thus efficiently implement key-value stores. The irregular but self-similar nature of the network facilitates constructing large networks easily using prewired, commodity racks. We show that Small-World Datacenters can achieve higher bandwidth and fault tolerance compared to both conventional hierarchical datacenters as well as the recently proposed CamCube topology. Coupled with hardware acceleration for packet switching, small-world datacenters can achieve an order of magnitude higher bandwidth than a conventional datacenter, depending on the network traffic.
Multi-tenant datacenters represent an extremely challenging networking environment. Tenants want the ability to migrate unmodified workloads from their enterprise networks to service provider datacenters, retaining the same networking configurations of their home network. The service providers must meet these needs without operator intervention while preserving their own operational flexibility and efficiency. Traditional networking approaches have failed to meet these tenant and provider requirements. Responding to this need, we present the design and implementation of a network virtualization solution for multi-tenant datacenters.
Nowadays, more and more companies migrate business from their own servers to the cloud. With the influx of computational requests, datacenters consume tremendous energy every day, attracting great attention in the energy efficiency dilemma. In this paper, we investigate the energy-aware resource management problem in cloud datacenters, where green energy with unpredictable capacity is connected. Via proposing a robust blockchain-based decentralized resource management framework, we save the energy consumed by the request scheduler. Moreover, we propose a reinforcement learning method embedded in a smart contract to further minimize the energy cost. Because the reinforcement learning method is informed from the historical knowledge, it relies on no request arrival and energy supply. Experimental results on Google cluster traces and real-world electricity price show that our approach is able to reduce the datacenters cost significantly compared with other benchmark algorithms.
Reducing energy consumption in datacenters is key to building low cost datacenters. To address this challenge, we explore the potential of hybrid datacenter designs that mix low power platforms with high performance ones. We show how these designs can handle diverse workloads with different service level agreements in an energy efficient fashion. We evaluate the feasibility of our approach through experiments and then discuss the design challenges and options of hybrid datacenters.
The amount of CO <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$_{2}$</tex-math></inline-formula> emitted per kilowatt-hour on an electricity grid varies by time of day and substantially varies by location due to the types of generation. Networked collections of warehouse scale computers, sometimes called Hyperscale Computing, emit more carbon than needed if operated without regard to these variations in carbon intensity. This paper introduces Google’s system for global Carbon-Intelligent Compute Management, which actively minimizes electricity-based carbon footprint and power infrastructure costs by delaying temporally flexible workloads. The core component of the system is a suite of analytical pipelines used to gather the next day’s carbon intensity forecasts, train day-ahead demand prediction models, and use risk-aware optimization to generate the next day’s carbon-aware Virtual Capacity Curves (VCCs) for all datacenter clusters across Google’s fleet. VCCs impose hourly limits on resources available to temporally flexible workloads while preserving overall daily capacity, enabling all such workloads to complete within a day with high probability. Data from Google’s in-production operation shows that VCCs effectively limit hourly capacity when the grid’s energy supply mix is carbon intensive and delay the execution of temporally flexible workloads to “greener” times.
Pervasive use of cloud computing and the resulting rise in the number of datacenters and hosting centers (that provide platform or software services to clients who do not have the means to set up and operate their own computing facilities) have brought forth many concerns, including the electrical energy cost, peak power dissipation, cooling, and carbon emission. With power consumption becoming an increasingly important issue for the operation and maintenance of the hosting centers, corporate and business owners are becoming increasingly concerned. Furthermore, provisioning resources in a cost-optimal manner so as to meet different performance criteria, such as throughput or response time, has become a critical challenge. The goal of this paper is to provide an introduction to resource provisioning and power or thermal management problems in datacenters, and to review strategies that maximize the datacenter energy efficiency subject to peak or total power consumption and thermal constraints, while meeting stipulated service level agreements in terms of task throughput and/or response time.
Small RTTs (~tens of microseconds), bursty flow arrivals, and a large number of concurrent flows (thousands) in datacenters bring fundamental challenges to congestion control as they either force a flow to send at most one packet per RTT or induce a large queue build-up. The widespread use of shallow buffered switches also makes the problem more challenging with hosts generating many flows in bursts. In addition, as link speeds increase, algorithms that gradually probe for bandwidth take a long time to reach the fair-share. An ideal datacenter congestion control must provide 1) zero data loss, 2) fast convergence, 3) low buffer occupancy, and 4) high utilization. However, these requirements present conflicting goals.
Photonic switches are increasingly considered for insertion in high performance datacenter architectures to meet the growing performance demands of interconnection networks. We provide an overview of photonic switching technologies and develop an evaluation methodology for assessing their potential impact on datacenter performance. We begin with a review of three categories of optical switches, namely, free-space switches, III-V integrated switches and silicon integrated switches. The state-of-the-art of MEMS, LCOS, SOA, MZI and MRR switching technologies are covered, together with insights on their performance limitations and scalability considerations. The performance metrics that are required for optical switches to truly emerge in datacenters are discussed and summarized, with special focus on the switching time, cost, power consumption, scalability and optical power penalty. Furthermore, the Pareto front of the switch metric space is analyzed. Finally, we propose a hybrid integrated switch fabric design using the III-V/Si wafer bonding technique and investigate its potential impact on realizing reduced cost and power penalty.
Datacenter power consumption has a significant impact on both its recurring electricity bill (Op-ex) and one-time construction costs (Cap-ex). Existing work optimizing these costs has relied primarily on throttling devices or workload shaping, both with performance degrading implications. In this paper, we present a novel knob of energy buffer (eBuff) available in the form of UPS batteries in datacenters for this cost optimization. Intuitively, eBuff stores energy in UPS batteries during "valleys" - periods of lower demand, which can be drained during "peaks" - periods of higher demand. UPS batteries are normally used as a fail-over mechanism to transition to captive power sources upon utility failure. Furthermore, frequent discharges can cause UPS batteries to fail prematurely. We conduct detailed analysis of battery operation to figure out feasible operating regions given such battery lifetime and datacenter availability concerns. Using insights learned from this analysis, we develop peak reduction algorithms that combine the UPS battery knob with existing throttling based techniques for minimizing datacenter power costs. Using an experimental platform, we offer insights about Op-ex savings offered by eBuff for a wide range of workload peaks/valleys, UPS provisioning, and application SLA constraints. We find that eBuff can be used to realize 15-45% peak power reduction, corresponding to 6-18% savings in Op-ex across this spectrum. eBuff can also play a role in reducing Cap-ex costs by allowing tighter overbooking of power infrastructure components and we quantify the extent of such Cap-ex savings. To our knowledge, this is the first paper to exploit stored energy - typically lying untapped in the datacenter - to address the peak power draw problem.
Several application domains are collecting data using Internet of Things sensing devices and shipping it to remote cloud datacenters for analysis (fusion, storage, and processing). Data analytics activities raise a new set of technical challenges from the perspective of ensuring end-to-end security and privacy of data as it travels from an edge datacenter (EDC) to a cloud datacenter (CDC) (or vice versa). This article discusses the security threats in EDCs and CDCs by dividing the complete network structure into three layers: perception layer, network layer, and application layer.
In spite of various gains, cloud computing has got few challenges and issues including dynamic resource scaling and power consumption. Such affairs cause a cloud system to be fragile and expensive. In this paper we address both issues in cloud datacenter through workload prediction. The workload prediction model is developed using long short term memory (LSTM) networks. The proposed model is tested on three benchmark datasets of web server logs. The empirical results show that the proposed method achieved high accuracy in predictions by reducing the mean squared error up to 3.17 x 10-3.
暂无摘要（点击查看原文获取完整内容）
Business-critical workloads -- web servers, mail servers, app servers, etc. -- are increasingly hosted in virtualized data enters acting as Infrastructure-as-a-Service clouds (cloud data enters). Understanding how business-critical workloads demand and use resources is key in capacity sizing, in infrastructure operation and testing, and in application performance management. However, relatively little is currently known about these workloads, because the information is complex -- larges-scale, heterogeneous, shared-clusters -- and because datacenter operators remain reluctant to share such information. Moreover, the few operators that have shared data (e.g., Google and several supercomputing centers) have enabled studies in business intelligence (MapReduce), search, and scientific computing (HPC), but not in business-critical workloads. To alleviate this situation, in this work we conduct a comprehensive study of business-critical workloads hosted in cloud data enters. We collect two large-scale and long-term workload traces corresponding to requested and actually used resources in a distributed datacenter servicing business-critical workloads. We perform an in-depth analysis about workload traces. Our study sheds light into the workload of cloud data enters hosting business-critical workloads. The results of this work can be used as a basis to develop efficient resource management mechanisms for data enters. Moreover, the traces we released in this work can be used for workload verification, modelling and for evaluating resource scheduling policies, etc.
As a key component in a modern datacenter, the cloud operating system is responsible for managing the physical and virtual infrastructure, orchestrating and commanding service provisioning and deployment, and providing federation capabilities for accessing and deploying virtual resources in remote cloud infrastructures.
Energy storage - in the form of UPS units - in a datacenter has been primarily used to fail-over to diesel generators upon power outages. There has been recent interest in using these Energy Storage Devices (ESDs) for demand-response (DR) to either shift peak demand away from high tariff periods, or to shave demand allowing aggressive under-provisioning of the power infrastructure. All such prior work has only considered a single/specific type of ESD (typically re-chargeable lead-acid batteries), and has only employed them at a single level of the power delivery network. Continuing technological advances have provided us a plethora of competitive ESD options ranging from ultra-capacitors, to different kinds of batteries, flywheels and even compressed air-based storage. These ESDs offer very different trade-offs between their power and energy costs, densities, lifetimes, and energy efficiency, among other factors, suggesting that employing hybrid combinations of these may allow more effective DR than with a single technology. Furthermore, ESDs can be placed at different, and possibly multiple, levels of the power delivery hierarchy with different associated trade-offs. To our knowledge, no prior work has studied the extensive design space involving multiple ESD technology provisioning and placement options. This paper intends to fill this critical void, by presenting a theoretical framework for capturing important characteristics of different ESD technologies, the trade-offs of placing them at different levels of the power hierarchy, and quantifying the resulting cost-benefit trade-offs as a function of workload properties.
Datacenter networks have been designed to tolerate failures of network equipment and provide sufficient bandwidth. In practice, however, failures and maintenance of networking and power equipment often make tens to thousands of servers unavailable, and network congestion can increase service latency. Unfortunately, there exists an inherent tradeoff between achieving high fault tolerance and reducing bandwidth usage in network core; spreading servers across fault domains improves fault tolerance, but requires additional bandwidth, while deploying servers together reduces bandwidth usage, but also decreases fault tolerance. We present a detailed analysis of a large-scale Web application and its communication patterns. Based on that, we propose and evaluate a novel optimization framework that achieves both high fault tolerance and significantly reduces bandwidth usage in the network core by exploiting the skewness in the observed communication patterns.
Managing data and computation is at the heart of datacenter computing. Manual management of data can lead to data loss, wasteful consumption of storage, and laborious bookkeeping. Lack of proper management of computation can result in lost opportunities to share common computations across multiple jobs or to compute results incrementally. Nectar is a system designed to address the aforementioned problems. It automates and unifies the management of data and computation within a datacenter. In Nectar, data and computation are treated interchangeably by associating data with its computation. Derived datasets, which are the results of computations, are uniquely identified by the programs that produce them, and together with their programs, are automatically managed by a datacenter wide caching service. Any derived dataset can be transparently regenerated by re-executing its program, and any computation can be transparently avoided by using previously cached results. This enables us to greatly improve datacenter management and resource utilization: obsolete or infrequently used derived datasets are automatically garbage collected, and shared common computations are computed only once and reused by others. This paper describes the design and implementation of Nectar, and reports on our evaluation of the system using analytic studies of logs from several production clusters and an actual deployment on a 240-node cluster. 1
We introduce a new and increasingly relevant setting for distributed optimization in machine learning, where the data defining the optimization are distributed (unevenly) over an extremely large number of \nodes, but the goal remains to train a high-quality centralized model. We refer to this setting as Federated Optimization. In this setting, communication efficiency is of utmost importance. A motivating example for federated optimization arises when we keep the training data locally on users' mobile devices rather than logging it to a data center for training. Instead, the mobile devices are used as nodes performing computation on their local data in order to update a global model. We suppose that we have an extremely large number of devices in our network, each of which has only a tiny fraction of data available totally; in particular, we expect the number of data points available locally to be much smaller than the number of devices. Additionally, since different users generate data with different patterns, we assume that no device has a representative sample of the overall distribution. We show that existing algorithms are not suitable for this setting, and propose a new algorithm which shows encouraging experimental results. This work also sets a path for future research needed in the context of federated optimization.