Deep learning (DL) system research is often impeded by the limited availability and expensive costs of GPUs. In this paper, we introduce GPEmu, a GPU emulator for faster and cheaper prototyping and evaluation of deep learning system research without using real GPUs. GPEmu comes with four novel features: time emulation, memory emulation, distributed system support, and sharing support. We support over 30 DL models and 6 GPU models, the largest scale to date. We demonstrate the power of GPEmu by successfully reproducing the main results of nine recent publications and easily prototyping three new micro-optimizations.

We present Dinomo, a novel key-value store for disaggregated persistent memory (DPM). Dinomo is the first key-value store for DPM that simultaneously achieves high common-case performance, scalability, and lightweight online reconfiguration. We observe that previously proposed key-value stores for DPM had architectural limitations that prevent them from achieving all three goals simultaneously. Dinomo uses a novel combination of techniques such as ownership partitioning, disaggregated adaptive caching, selective replication, and lock-free and log-free indexing to achieve these goals. Compared to a state-of-the-art DPM key-value store, Dinomo achieves at least 3.8X better throughput at scale on various workloads and higher scalability, while providing fast reconfiguration.

Most modern data processing pipelines run on top of a distributed storage layer, and securing the whole system, and the storage layer in particular, against accidental or malicious misuse is crucial to ensuring compliance to rules and regulations. Enforcing data protection and privacy rules, however, stands at odds with the requirement to achieve higher and higher access bandwidths and processing rates in large data processing pipelines. In this work we describe our proposal for the path forward that reconciles the two goals. We call our approach "Software-Defined Data Protection" (SDP). Its premise is simple, yet powerful: decoupling often changing policies from request-level enforcement allows distributed smart storage nodes to implement the latter at line-rate. Existing and future data protection frameworks can be translated to the same hardware interface which allows storage nodes to offload enforcement efficiently both for company-specific rules and regulations, such as GDPR or CCPA. While SDP is a promising approach, there are several remaining challenges to making this vision reality. As we explain in the paper, overcoming these will require collaboration across several domains, including security, databases and specialized hardware design.

Training Deep Neural Networks (DNNs) is resource-intensive and time-consuming. While prior research has explored many different ways of reducing DNN training time, the impact of input data pipeline , i.e., fetching raw data items from storage and performing data pre-processing in memory, has been relatively unexplored. This paper makes the following contributions: (1) We present the first comprehensive analysis of how the input data pipeline affects the training time of widely-used computer vision and audio Deep Neural Networks (DNNs), that typically involve complex data pre-processing. We analyze nine different models across three tasks and four datasets while varying factors such as the amount of memory, number of CPU threads, storage device, GPU generation etc on servers that are a part of a large production cluster at Microsoft. We find that in many cases, DNN training time is dominated by data stall time : time spent waiting for data to be fetched and pre-processed. (2) We build a tool, DS-Analyzer to precisely measure data stalls using a differential technique, and perform predictive what-if analysis on data stalls. (3) Finally, based on the insights from our analysis, we design and implement three simple but effective techniques in a data-loading library, CoorDL, to mitigate data stalls. Our experiments on a range of DNN tasks, models, datasets, and hardware configs show that when PyTorch uses CoorDL instead of the state-of-the-art DALI data loading library, DNN training time is reduced significantly (by as much as 5X on a single server).

The General Data Protection Regulation (GDPR) provides new rights and protections to European people concerning their personal data. We analyze GDPR from a systems perspective, translating its legal articles into a set of capabilities and characteristics that compliant systems must support. Our analysis reveals the phenomenon of metadata explosion, wherein large quantities of metadata needs to be stored along with the personal data to satisfy the GDPR requirements. Our analysis also helps us identify new workloads that must be supported under GDPR. We design and implement an open-source benchmark called GDPRbench that consists of workloads and metrics needed to understand and assess personal-data processing database systems. To gauge the readiness of modern database systems for GDPR, we follow best practices and developer recommendations to modify Redis, PostgreSQL, and a commercial database system to be GDPR compliant. Our experiments demonstrate that the resulting GDPR-compliant systems achieve poor performance on GPDR workloads, and that performance scales poorly as the volume of personal data increases. We discuss the real-world implications of these .ndings, and identify research challenges towards making GDPR-compliance efficient in production environments. We release all of our so.ware artifacts and datasets at h.p://www:gdprbench:org