Publications

Achieving high performance in machine learning workloads is a crucial yet difficult task. To achieve high runtime performance on hardware platforms such as GPUs, graph-based executions such as CUDA graphs are often used to eliminate CPU runtime overheads by submitting jobs in the granularity of multiple kernels. However, many machine learning workloads, especially dynamic deep neural networks (DNNs) with

Nowadays, VPN technology is widely used in cloud and hybrid network communication that makes use of algorithms and tunneling to meet different security requirements. However, existing cloud VPN gateways often lack advanced monitoring capabilities and struggle to identify and resolve network connectivity and performance issues. Hence, LPMLP adapted Secure cloud VPN Gateway with Network Monitoring and Issue

Graph sampling prepares training samples for graph learning and can dominate the training time. Due to the increasing algorithm diversity and complexity, existing sampling frameworks are insufficient in the generality of expression and the efficiency of execution. To close this gap, we conduct a comprehensive study on 15 popular graph sampling algorithms to motivate the design of gSampler, a general and

Verification of asynchronous distributed programs is challenging due to the need to reason about numerous control paths resulting from the myriad interleaving of messages and failures. In this paper, we propose an automated bookkeeping method based on message chains. Message chains reveal structure in asynchronous distributed system executions and can help programmers verify their systems at the message

Automated static dataflow analysis is an effective technique for detecting security critical issues like sensitive data leak, and vulnerability to injection attacks. Ensuring high precision and recall requires an analysis that is context, field and object sensitive. However, it is challenging to attain high precision and recall and scale to large industrial code bases. Compositional style analyses in which

Time-to-insight is a critical measure in a number of satellite mission applications: detection and warning of fastmoving events like fires and floods, or identification and tracking of satellites or missiles, for example. Current data flows delay the time-to-insight on the order of minutes or hours, as all collected data must be downlinked in one or more contact windows, then transited over terrestrial

AWS Glue is Amazon’s serverless data integration cloud service that makes it simple and cost effective to extract, clean, enrich, load, and organize data. Originally launched in August 2017, AWS Glue began as an extract-transform-load (ETL) service designed to relieve developers and data engineers of the undifferentiated heavy lifting needed to load databases, data warehouses, and build data lakes on Amazon

Federated learning (FL) enables multiple client devices to train a single machine learning model collaboratively. As FL often involves various smart devices, it is important to adapt the FL pipeline to accommodate device resource constraints. This work addresses the problem of training and storing memory-intensive deep neural network architectures on resource-constrained devices. Existing solutions often

Reliable storage systems must be crash consistent – guaranteed to recover to a consistent state after a crash. Crash consistency is non-trivial as it requires maintaining complex invariants about persistent data structures in the presence of caching, reordering, and system failures. Current programming models offer little support for implementing crash consistency, forcing storage system developers to roll

Fulfillment centers in the E-commerce industry are highly complex systems that houses inventory and fulfill customer orders. One of the key processes at these centers involves translating customer demands into trucks and yard operations. Truck yards with operational issues can create delays in customer orders. In this paper, we show how a scalable cloud-based hybrid simulation model is used to improve yard

AWS Lambda is a serverless event-driven compute service, part of a category of cloud compute offerings sometimes called Function-as-a-service (FaaS). When we first released AWS Lambda, functions were limited to 250MB of code and dependencies, packaged as a simple compressed archive. In 2020, we released support for deploying container images as large as 10GiB as Lambda functions, allowing customers to bring

There has been a lot of excitement around using machine learning to improve the performance and usability of database systems. However, few of these techniques have actually been used in the critical path of customer-facing database services. In this paper, we describe Auto-WLM, a machine learning based automatic workload manager currently used in production in Amazon Redshift. Auto-WLM is an example of

This paper proposes a new Electromagnetic BandGap (EBG) Structure based filter design. This on-PCB filter imitates the functionality of a discrete low-pass filter without any of the associated Bill of Materials (BOM) costs. The methodology of constructing the EBG filter is explained in detail, making use of classical filter theory including tapering and multiple elements that enable tradeoffs between roll-off

In live video streaming, the size of Dynamic Adaptive Streaming over HTTP (DASH) manifest grows as the number of periods increases and/or the overall time duration of DASH manifest increases. The bigger the manifest size in bytes is, the more computation for manifest generation, manifest storage and parsing and network traffic there will be on service side, the more data to be downloaded, manifest refresh

Serverless cloud-based warehousing systems enable users to create materialized views in order to speed up predictable and repeated query workloads. Incremental view maintenance (IVM) minimizes the time needed to bring a materialized view up-to-date. It allows the refresh of a materialized view solely based on the base table changes since the last refresh. In serverless cloud-based warehouses, IVM uses computations