Data and code working together is fundamental to machine learning (ML), but the context around datasets and interactions between datasets and code are in general captured only rudimentarily. Context such as how the dataset was prepared and created, what source data were used, what code was used in processing, how the dataset evolved, and where it has been used and reused can provide much insight, but this information is often poorly documented. That is unfortunate since it makes datasets into black-boxes with potentially hidden characteristics that have downstream consequences. We argue that making dataset preparation more accessible and dataset usage easier to record and document would have significant benefits for the ML community: it would allow for greater diversity in datasets by inviting modification to published sources, simplify use of alternative datasets and, in doing so, make results more transparent and robust, while allowing for all contributions to be adequately credited. We present a platform, Renku, designed to support and encourage such sustainable development and use of data, datasets, and code, and we demonstrate its benefits through a few illustrative projects which span the spectrum from dataset creation to dataset consumption and showcasing.

Emerging scale-out workloads require extensive amounts of computational resources. However, data centers using modern server hardware face physical constraints in space and power, limiting further expansion and calling for improvements in the computational density per server and in the per-operation energy. Continuing to improve the computational resources of the cloud while staying within physical constraints mandates optimizing server efficiency to ensure that server hardware closely matches the needs of scale-out workloads.

Chip-multiprocessors require a coherence directory to track data sharing and order accesses to the shared data. Scaling coherence directories to support a large number of cores is challenging due to excessive area requirements of the directories. The state-of-the-art proposals reduce the directory size by not keeping coherence information for private data. These approaches are useful for workloads that have predominantly private data, but are not applicable to workloads with shared data. We observe that data are not actively shared by multiple cores. In workloads with a shared dataset, although each core accesses the whole data, the chance that multiple cores access the same piece of data at the same time is low. Based on this observation we design a Spatiotemporal Coherence Tracking scheme that drastically reduces the directory size without sacrificing performance. The proposed directory scheme uses dual-grain tracking and switches between the granularities whenever possible to save the area. It dynamically detects spatial regions of data that are privately accessed by one core over a time period and for those regions, increases coherence tracking granularity from block-level to region-level. Our experimental results show that the proposed approach can reduce the baseline sparse directory size by at least 75% across a variety of commercial and scientific workloads, while sacrificing only 1% of performance. Using our approach, the directory can be under-provisioned to have fewer entries than the number of cache blocks that are being tracked.

Emerging scale-out workloads require extensive amounts of computational resources. However, data centers using modern server hardware face physical constraints in space and power, limiting further expansion and calling for improvements in the computational density per server and in the per-operation energy. Continuing to improve the computational resources of the cloud while staying within physical constraints mandates optimizing server efficiency to ensure that server hardware closely matches the needs of scale-out workloads. In this work, we introduce CloudSuite, a benchmark suite of emerging scale-out workloads. We use performance counters on modern servers to study scale-out workloads, finding that today’s predominant processor microarchitecture is inefficient for running these workloads. We find that inefficiency comes from the mismatch between the workload needs and modern processors, particularly in the organization of instruction and data memory systems and the processor core microarchitecture. Moreover, while today’s predominant microarchitecture is inefficient when executing scale-out workloads, we find that continuing the current trends will further exacerbate the inefficiency in the future. In this work, we identify the key microarchitectural needs of scale-out workloads, calling for a change in the trajectory of server processors that would lead to improved computational density and power efficiency in data centers.