arXiv subjects

Lalith Suresh

Publications and source records attributed to Lalith Suresh.

8 recordsLinked to original sources

arXiv2023

OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud

We present OptiReduce, a new collective-communication system for the cloud with bounded, predictable completion times for deep-learning jobs in the presence of varying computation (stragglers) and communication (congestion and gradient drops) variabilities. OptiReduce exploits the inherent resiliency and the stochastic nature of distributed deep-learning (DDL) training and fine-tuning to work with approximated (or lost) gradients -- providing an efficient balance between (tail) performance and the resulting accuracy of the trained models. Exploiting this domain-specific characteristic of DDL, OptiReduce introduces (1) mechanisms (e.g., unreliable bounded transport with adaptive timeout) to improve the DDL jobs' tail execution time, and (2) strategies (e.g., Transpose AllReduce and Hadamard Transform) to mitigate the impact of gradient drops on model accuracy. Our evaluation shows that OptiReduce achieves 70% and 30% faster time-to-accuracy (TTA), on average, when operating in shared, cloud environments (e.g., CloudLab) compared to Gloo and NCCL, respectively.

cs.DC↗

arXiv2023

Apiary: A DBMS-Integrated Transactional Function-as-a-Service Framework

Developers increasingly use function-as-a-service (FaaS) platforms for data-centric applications that perform low-latency and transactional operations on data, such as for microservices or web serving. Unfortunately, existing FaaS platforms support these applications poorly because they physically and logically separate application logic, executed in cloud functions, from data management, done in interactive transactions accessing remote storage. Physical separation harms performance while logical separation complicates efficiently providing transactional guarantees and fault tolerance. This paper introduces Apiary, a novel DBMS-integrated FaaS platform for deploying and composing fault-tolerant transactional functions. Apiary physically co-locates and logically integrates function execution and data management by wrapping a distributed DBMS engine and using it as a unified runtime for function execution, data management, and operational logging, thus providing similar or stronger transactional guarantees as comparable systems while greatly improving performance and observability. To allow developers to write complex stateful programs, we leverage this integration to enable efficient and fault-tolerant function composition, building a frontend for orchestrating workflows of functions with the guarantees that each workflow runs to completion and each function in a workflow executes exactly once. We evaluate Apiary against research and production FaaS platforms and show it outperforms them by 2--68x on microservice workloads by reducing communication overhead.

cs.DB↗

arXiv2022

Transactions Make Debugging Easy

We propose TROD, a novel transaction-oriented framework for debugging modern distributed web applications and online services. Our critical insight is that if applications store all state in databases and only access state transactionally, TROD can use lightweight always-on tracing to track the history of application state changes and data provenance, and then leverage the captured traces and transaction logs to faithfully replay or even test modified code retroactively on any past event. We demonstrate how TROD can simplify programming and debugging in production applications, list several research challenges and directions, and encourage the database and systems communities to drastically rethink the synergy between the way people develop and debug applications.

cs.DB↗

arXiv2021

PL2: Towards Predictable Low Latency in Rack-Scale Networks

High performance rack-scale offerings package disaggregated pools of compute, memory and storage hardware in a single rack to run diverse workloads with varying requirements, including applications that need low and predictable latency. The intra-rack network is typically high speed Ethernet, which can suffer from congestion leading to packet drops and may not satisfy the stringent tail latency requirements for some workloads (including remote memory/storage accesses). In this paper, we design a Predictable Low Latency(PL2) network architecture for rack-scale systems with Ethernet as interconnecting fabric. PL2 leverages programmable Ethernet switches to carefully schedule packets such that they incur no loss with NIC and switch queues maintained at small, near-zero levels. In our 100 Gbps rack-prototype, PL2 keeps 99th-percentile memcached RPC latencies under 60us even when the RPCs compete with extreme offered-loads of 400%, without losing traffic. Network transfers for a machine learning training task complete 30% faster than a receiver-driven scheme implementation modeled after Homa (222ms vs 321ms 99%ile latency per iteration).

cs.NI↗

arXiv2019

Automating Cluster Management with Weave

Modern cluster management systems like Kubernetes and Openstack grapple with hard combinatorial optimization problems: load balancing, placement, scheduling, and configuration. Currently, developers tackle these problems by designing custom application-specific algorithms---an approach that is proving unsustainable, as ad-hoc solutions both perform poorly and introduce overwhelming complexity to the system, making it challenging to add important new features. We propose a radically different architecture, where programmers drive cluster management tasks declaratively, using SQL queries over cluster state stored in a relational database. These queries capture in a natural way both constraints on the cluster configuration as well as optimization objectives. When a cluster reconfiguration is required at runtime, our tool, called Weave, synthesizes an encoding of these queries into an optimization model, which it solves using an off-the-shelf solver. We demonstrate Weave's efficacy by powering three production-grade systems with it: a Kubernetes scheduler, a virtual machine management solution, and a distributed transactional datastore. Using Weave, we expressed complex cluster management policies in under 20 lines of SQL, easily added new features to these existing systems, and significantly improved placement quality and convergence times.

cs.DC↗

arXiv2019

Hillview: A trillion-cell spreadsheet for big data

Hillview is a distributed spreadsheet for browsing very large datasets that cannot be handled by a single machine. As a spreadsheet, Hillview provides a high degree of interactivity that permits data analysts to explore information quickly along many dimensions while switching visualizations on a whim. To provide the required responsiveness, Hillview introduces visualization sketches, or vizketches, as a simple idea to produce compact data visualizations. Vizketches combine algorithmic techniques for data summarization with computer graphics principles for efficient rendering. While simple, vizketches are effective at scaling the spreadsheet by parallelizing computation, reducing communication, providing progressive visualizations, and offering precise accuracy guarantees. Using Hillview running on eight servers, we can navigate and visualize datasets of tens of billions of rows and trillions of cells, much beyond the published capabilities of competing systems.

cs.DC↗

arXiv2018

Elmo: Source-Routed Multicast for Cloud Services

We present Elmo, a system that addresses the multicast scalability problem in multi-tenant data centers. Modern cloud applications frequently exhibit one-to-many communication patterns and, at the same time, require sub-millisecond latencies and high throughput. IP multicast can achieve these requirements but has control- and data-plane scalability limitations that make it challenging to offer it as a service for hundreds of thousands of tenants, typical of cloud environments. Tenants, therefore, must rely on unicast-based approaches (e.g., application-layer or overlay-based) to support multicast in their applications, imposing overhead on throughput and end host CPU utilization, with higher and unpredictable latencies. Elmo scales network multicast by taking advantage of emerging programmable switches and the unique characteristics of data-center networks; specifically, the symmetric topology and short paths in a data center. Elmo encodes multicast group information inside packets themselves, reducing the need to store the same information in network switches. In a three-tier data-center topology with 27K hosts, Elmo supports a million multicast groups using a 325-byte packet header, requiring as few as 1.1K multicast group-table entries on average in leaf switches, with a traffic overhead as low as 5% over ideal multicast.

cs.NI↗

arXiv2018

Stable and Consistent Membership at Scale with Rapid

We present the design and evaluation of Rapid, a distributed membership service. At Rapid's core is a scheme for multi-process cut detection (CD) that revolves around two key insights: (i) it suspects a failure of a process only after alerts arrive from multiple sources, and (ii) when a group of processes experience problems, it detects failures of the entire group, rather than conclude about each process individually. Implementing these insights translates into a simple membership algorithm with low communication overhead. We present evidence that our strategy suffices to drive unanimous detection almost-everywhere, even when complex network conditions arise, such as one-way reachability problems, firewall misconfigurations, and high packet loss. Furthermore, we present both empirical evidence and analyses that proves that the almost-everywhere detection happens with high probability. To complete the design, Rapid contains a leaderless consensus protocol that converts multi-process cut detections into a view-change decision. The resulting membership service works both in fully decentralized as well as logically centralized modes. We present an evaluation of Rapid in moderately scalable cloud settings. Rapid bootstraps 2000 node clusters 2-5.8x faster than prevailing tools such as Memberlist and ZooKeeper, remains stable in face of complex failure scenarios, and provides strong consistency guarantees. It is easy to integrate Rapid into existing distributed applications, of which we demonstrate two.

cs.DC↗