Geo-distributed Data Analytics Research Articles

Multi-Stage Geo-Distributed Data Aggregation With Coordinated Computation and Communication in Edge Compute First Networking

With the explosion of geo-distributed data, the huge treasures hidden in them are waiting to be explored to obtain valuable insights. This results in the need for an effective geo- distributed data analysis method. The traditional approach to geo-distributed data analytics is to gather all the required data into a single edge datacenter (edge DC) through one transmission and aggregation (centralized data aggregation). However, as the volume of data grows exponentially, the centralized data aggregation scheme becomes inefficient or infeasible due to the limitations of the computing and network resources. In this paper, we propose the geo-distributed data aggregation scheme in edge compute first networking (CFN) with joint consideration of computation and communication resources. The proposed scheme optimizes two objectives: the first is to minimize the job completion time (JCT) by selecting cluster centers, dividing clusters and provisioning lightpaths; the second objective is to reduce bandwidth consumption by reallocating routing and frequency slots based on JCT. To achieve the two objectives, we first formulate the optimization problem of multi-stage geo- distributed data aggregation as a linear programming (LP) model. To tackle the computational complexity issue of the LP model, a multi-stage geo-distributed data aggregation algorithm jointly with computation and communication resources (MGDD-CC) is proposed. Simulation results show that the proposed scheme can reduce JCT, alleviate the competition for bandwidth resources and is more suitable for scenarios with better data aggregation effects and larger quantities of geo-distributed data.

Geo-Distributed IoT Data Analytics With Deadline Constraints Across Network Edge

Owing to the advancement of the Internet of Things (IoT) and 5G mobile technologies, various IoT devices produce massive data, which is usually transferred to nearby sites, such as edge nodes or datacenters. Many large-scale IoT applications need to analyze the data distributed across multiple sites to obtain final results. A dominant challenge of this type of data analytics is the heterogeneities of resource capacities across geo-distributed sites. In this article, we find that the resource capacity as well as the resource price differ among sites, and the price heterogeneity has a significant impact on geo-distributed IoT data analytics. Thus, each geo-distributed IoT data analytics job prefers to minimize the job execution cost while guaranteeing its deadline requirement under the resource constraints of involved sites. Specifically, we propose to jointly consider the resource heterogeneities of both capacity and price, and minimize the cost of each job before its deadline. We characterize this optimization problem as a quadratically constrained quadratic programming problem. To tackle such an NP-hard problem, we propose the minimize the job completion cost before a given deadline (MCGL) method, which calculates a task placement solution by the gradient adjustment strategy according to the remarkable negative correlation relationship between job completion time and job completion cost of geo-distributed IoT data analytics job. The task placement strategy can optimize resource cost with respect to the deadline requirement of any geo-distributed data analytics job. The trace-driven evaluations indicate that MCGL significantly reduces the total cost compared with existing methods; moreover, they satisfy the deadline constraints simultaneously.

Network Cost-Aware Geo-Distributed Data Analytics System

Many geo-distributed data analytics (GDA) systems have focused on the network performance-bottleneck: inter-data center network bandwidth to improve performance. Unfortunately, these systems may encounter a <i>cost-bottleneck</i> ( <inline-formula><tex-math notation="LaTeX">${\$}$</tex-math></inline-formula> ) because they have not considered data transfer cost ( <inline-formula><tex-math notation="LaTeX">${\$}$</tex-math></inline-formula> ), one of the most expensive and heterogeneous resources in a multi-cloud environment. In this article, we present <i>Kimchi</i> , a network cost-aware GDA system to meet the cost-performance tradeoff by exploiting data transfer cost heterogeneity to avoid the cost-bottleneck. Kimchi determines cost-aware task placement decisions for scheduling tasks given inputs including data transfer cost, network bandwidth, input data size and locations, and desired cost-performance tradeoff preference. In addition, Kimchi is also mindful of data transfer cost in the presence of dynamics. Kimchi has been applied to two common GDA MapReduce models: synchronous barrier and asynchronous push-based shuffle. A Kimchi prototype has been implemented on Spark, and experiments show that it reduces cost by 5% <inline-formula><tex-math notation="LaTeX">$\scriptstyle \sim$</tex-math></inline-formula> 24% without impacting performance and reduces query execution time by 45% <inline-formula><tex-math notation="LaTeX">$\scriptstyle \sim$</tex-math></inline-formula> 70% without impacting cost compared to other baseline approaches centralized, vanilla Spark, and bandwidth-aware (e.g., Iridium). More importantly, Kimchi allows applications to explore a much richer cost-performance tradeoff space in a multi-cloud environment.

Trading Cost and Throughput in Geo-Distributed Analytics With A Two Time Scale Approach

In the era of global-scale services, analytical queries are performed on datasets that span multiple data centers (DCs). Such geo-distributed queries generate a large amount of inter-DC data transfers at run time. Due to the expensive inter-DC bandwidth, various methods have been proposed to reduce the traffic cost in geo-distributed data analytics. However, current methods do not attempt to address the throughput issue in geo-distributed analytics. In this article, we target at characterizing and optimizing a cost-throughput tradeoff problem in geo-distributed data analytics. Our objectives are two-fold: (1) we minimize the inter-DC traffic cost when serving geo-distributed analytics with uncertain query demand, and (2) we maximize the system throughput, in terms of the number of query requests that can be successfully served with guaranteed queuing delay. Specifically, we formulate a stochastic optimization problem that seamlessly combines these two objectives. To solve this problem, we take advantage of Lyapunov optimization techniques to design and analyze a two-timescale online control framework. Without prior knowledge of future query requests, this framework makes online decisions on input data placement and admission control of query requests. Rigorous theoretical analyses show that our framework can achieve a near-optimal solution and maintain system stability and robustness as well. Extensive trace-driven simulation results further demonstrate that our framework is capable of reducing inter-DC traffic cost, improving system throughput, and guaranteeing a maximum delay for each query request.

Optimizing Geo-Distributed Data Analytics with Coordinated Task Scheduling and Routing

Recent trends show that cloud computing is growing to span more and more globally distributed datacenters. For geo-distributed datacenters, there is an increasingly need for scheduling algorithms to place tasks across datacenters, by jointly considering WAN traffic and computation. This scheduling must deal with situations such as wide-area distributed data, data sharing, WAN bandwidth costs and datacenter capacity limits, while also minimizing makespan. However, this scheduling problem is NP-hard. We propose a new resource allocation algorithm called HPS + , an extension to Hypergraph Partition-based Scheduling. HPS + models the combined task-data dependencies and data-datacenter dependencies as an augmented hypergraph, and adopts an improved hypergraph partition technique to minimize WAN traffic. It further uses a coordination mechanism to allocate network resources closely following the guidelines of task requirements, for minimizing the makespan. Evaluation across the real China-Astronomy-Cloud model and Google datacenter model show that HPS + saves the amount of data transfers by upto 53 percent and reduces the makespan by 39 percent compared to existing algorithms.

Cost-Minimizing Online Algorithms for Geo-Distributed Data Analytics

Modern enterprises often manage geographically distributed datacenters around the globe. In such environment, datasets are naturally collected and stored in different data centers and were later queried for complex analytics. In this paper, we study the Wide-Area Data Analytics problem, which aims to efficiently control data movements and achieve low latency for overall queries processing, both constrained by limited and expensive network resources across datacenters. Previous papers focus on offline settings of single analytical queries and do not consider time in optimizing system performance, and therefore ignores the dynamics of data and task placement in terms of inter-DC bandwidth utilization. In this paper, we consider the online setting and formulate a cost-minimizing optimization problem over time for arbitrary Directed Acyclic Graph query processing. Considering dynamics of network resource usage, we developed two online algorithms, Online Switch Resist (OSR) and Most Fixed Horizon Control (MFHC) with good competitive ratios. We performed extensive simulations and comparative studies using the TPC-CH benchmark and verified the efficacy of proposed algorithms. The algorithm we proposed is better than the existing algorithm, and its performance approximates the theoretical optimal value.

Unicorn: Unified resource orchestration for multi-domain, geo-distributed data analytics

As the data volume increases exponentially over time, data-intensive analytics benefits substantially from multi-organizational, geographically-distributed, collaborative computing, where different organizations contribute various yet scarce resources, e.g., computation, storage and networking resources, to collaboratively collect, share and analyze extremely large amounts of data. By analyzing the data analytics trace from the Compact Muon Solenoid (CMS) experiment, one of the largest scientific experiments in the world, and systematically examining the design of existing resource management systems for clusters, we show that the multi-domain, geo-distributed, resource-disaggregated nature of this new paradigm calls for a framework to manage a large set of distributively-owned, heterogeneous resources, with the objective of efficient resource utilization, following the autonomy and privacy of different domains, and that the fundamental challenge for designing such a framework is: how to accurately discover and represent resource availability of a large set of distributively-owned, heterogeneous resources across different domains with minimal information exposure from each domain? Existing resource management systems are designed for single-domain clusters and cannot address this challenge. In this paper, we design Unicorn, the first unified resource orchestration framework for multi-domain, geo-distributed data analytics. In Unicorn, we encode the resource availability for each domain into resource state abstraction, a variant of the network view abstraction extended to accurately represent the availability of multiple resources with minimal information exposure using a set of linear inequalities. We then design a novel, efficient cross-domain query algorithm and a privacy-preserving resource information integration protocol to discover and integrate the accurate, minimal resource availability information for a set of data analytics jobs across different domains. In addition, Unicorn also contains a global resource orchestrator that computes optimal resource allocation decisions for data analytics jobs. We implement a prototype of Unicorn and present preliminary evaluation results to demonstrate its efficiency and efficacy. We also give a full demonstration of the Unicorn system at SuperComputing 2017.

Low Latency Geo-distributed Data Analytics

Low latency analytics on geographically distributed datasets (across datacenters, edge clusters) is an upcoming and increasingly important challenge. The dominant approach of aggregating all the data to a single datacenter significantly inflates the timeliness of analytics. At the same time, running queries over geo-distributed inputs using the current intra-DC analytics frameworks also leads to high query response times because these frameworks cannot cope with the relatively low and variable capacity of WAN links. We present Iridium, a system for low latency geo-distributed analytics. Iridium achieves low query response times by optimizing placement of both data and tasks of the queries. The joint data and task placement optimization, however, is intractable. Therefore, Iridium uses an online heuristic to redistribute datasets among the sites prior to queries' arrivals, and places the tasks to reduce network bottlenecks during the query's execution. Finally, it also contains a knob to budget WAN usage. Evaluation across eight worldwide EC2 regions using production queries show that Iridium speeds up queries by 3× -- 19× and lowers WAN usage by 15% -- 64% compared to existing baselines.