Abstract-Tasks in modern data-parallel clusters have highly diverse resource requirements along CPU, memory, disk and network. We present Tetris, a multi-resource cluster scheduler that packs tasks to machines based on their requirements of all resource types. Doing so avoids resource fragmentation as well as over-allocation of the resources that are not explicitly allocated, both of which are drawbacks of current schedulers. Tetris adapts heuristics for the multidimensional bin packing problem to the context of cluster schedulers wherein task arrivals and machine availability change in an online manner and wherein task's resource needs change with time and with the machine that the task is placed at. In addition, Tetris improves average job completion time by preferentially serving jobs that have less remaining work. We observe that fair allocations do not o er the best performance and the above heuristics are compatible with a large class of fairness policies; hence, we show how to simultaneously achieve good performance and fairness. Tracedriven simulations and deployment of our Apache YARN prototype on a node cluster show gains of over in makespan and job completion time while achieving nearly perfect fairness.

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.

No abstract

No abstract

Resource-constrained IoT devices, such as sensors and actuators, have become ubiquitous in recent years. This has led to the generation of large quantities of data in real-time, which is an appealing target for AI systems. However, deploying machine learning models on such end-devices is nearly impossible. A typical solution involves offloading data to external computing systems (such as cloud servers) for further processing but this worsens latency, leads to increased communication costs, and adds to privacy concerns. To address this issue, efforts have been made to place additional computing devices at the edge of the network, i.e., close to the IoT devices where the data is generated. Deploying machine learning systems on such edge computing devices alleviates the above issues by allowing computations to be performed close to the data sources. This survey describes major research efforts where machine learning systems have been deployed at the edge of computer networks, focusing on the operational aspects including compression techniques, tools, frameworks, and hardware used in successful applications of intelligent edge systems.

No abstract

Edge computing is a trending notion introduced a decade ago as a new computing paradigm for interactive mobile applications. The initial vision of the edge was a multi-tenant resource that will be used opportunistically for low-latency mobile applications. Despite that vision, we see in practice a different set of applications, driven by large-scale enterprises that have emerged and are driving realworld edge deployments today. In these applications, the edge is the primary place of storage and computation and, if network conditions allow, the cloud is opportunistically used alongside. We show how these enterprise deployments are driving innovation in edge computing. Enterprise-driven scenarios have a different motivation for using the edge. Instead of latency, the primary factors are limited bandwidth and unreliability of the network link to the cloud. The enterprise deployment layout is also unique: on-premise, single-tenant edges with shared, redundant outbound links. These previously unexplored characteristics of enterprise-driven edge scenarios open up a number of unique and exciting future research challenges for our community.

No abstract