Anastasia Ailamaki

Clearing the clouds: a study of emerging scale-out workloads on modern hardware

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 todays predominant processor micro-architecture 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 micro-architecture. Moreover, while todays predominant micro-architecture 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 micro-architectural 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.

Reactive NUCA: near-optimal block placement and replication in distributed caches

Increases in on-chip communication delay and the large working sets of server and scientific workloads complicate the design of the on-chip last-level cache for multicore processors. The large working sets favor a shared cache design that maximizes the aggregate cache capacity and minimizes off-chip memory requests. At the same time, the growing on-chip communication delay favors core-private caches that replicate data to minimize delays on global wires. Recent hybrid proposals offer lower average latency than conventional designs, but they address the placement requirements of only a subset of the data accessed by the application, require complex lookup and coherence mechanisms that increase latency, or fail to scale to high core counts. In this work, we observe that the cache access patterns of a range of server and scientific workloads can be classified into distinct classes, where each class is amenable to different block placement policies. Based on this observation, we propose Reactive NUCA (R-NUCA), a distributed cache design which reacts to the class of each cache access and places blocks at the appropriate location in the cache. R-NUCA cooperates with the operating system to support intelligent placement, migration, and replication without the overhead of an explicit coherence mechanism for the on-chip last-level cache. In a range of server, scientific, and multiprogrammed workloads, R-NUCA matches the performance of the best cache design for each workload, improving performance by 14% on average over competing designs and by 32% at best, while achieving performance within 5% of an ideal cache design.

Reconstruction and Simulation of Neocortical Microcircuitry

UNLABELLED We present a first-draft digital reconstruction of the microcircuitry of somatosensory cortex of juvenile rat. The reconstruction uses cellular and synaptic organizing principles to algorithmically reconstruct detailed anatomy and physiology from sparse experimental data. An objective anatomical method defines a neocortical volume of 0.29 ± 0.01 mm(3) containing ~31,000 neurons, and patch-clamp studies identify 55 layer-specific morphological and 207 morpho-electrical neuron subtypes. When digitally reconstructed neurons are positioned in the volume and synapse formation is restricted to biological bouton densities and numbers of synapses per connection, their overlapping arbors form ~8 million connections with ~37 million synapses. Simulations reproduce an array of in vitro and in vivo experiments without parameter tuning. Additionally, we find a spectrum of network states with a sharp transition from synchronous to asynchronous activity, modulated by physiological mechanisms. The spectrum of network states, dynamically reconfigured around this transition, supports diverse information processing strategies. PAPERCLIP VIDEO ABSTRACT.

Toward Dark Silicon in Servers

Server chips will not scale beyond a few tens to low hundreds of cores, and an increasing fraction of the chip in future technologies will be dark silicon that we cannot afford to power. Specialized multicore processors, however, can leverage the underutilized die area to overcome the initial power barrier, delivering significantly higher performance for the same bandwidth and power envelopes.

Shore-MT: a scalable storage manager for the multicore era

Database storage managers have long been able to efficiently handle multiple concurrent requests. Until recently, however, a computer contained only a few single-core CPUs, and therefore only a few transactions could simultaneously access the storage managers internal structures. This allowed storage managers to use non-scalable approaches without any penalty. With the arrival of multicore chips, however, this situation is rapidly changing. More and more threads can run in parallel, stressing the internal scalability of the storage manager. Systems optimized for high performance at a limited number of cores are not assured similarly high performance at a higher core count, because unanticipated scalability obstacles arise. We benchmark four popular open-source storage managers (Shore, BerkeleyDB, MySQL, and PostgreSQL) on a modern multicore machine, and find that they all suffer in terms of scalability. We briefly examine the bottlenecks in the various storage engines. We then present Shore-MT, a multithreaded and highly scalable version of Shore which we developed by identifying and successively removing internal bottlenecks. When compared to other DBMS, Shore-MT exhibits superior scalability and 2--4 times higher absolute throughput than its peers. We also show that designers should favor scalability to single-thread performance, and highlight important principles for writing scalable storage engines, illustrated with real examples from the development of Shore-MT.

Data-oriented transaction execution

While hardware technology has undergone major advancements over the past decade, transaction processing systems have remained largely unchanged. The number of cores on a chip grows exponentially, following Moores Law, allowing for an ever-increasing number of transactions to execute in parallel. As the number of concurrently-executing transactions increases, contended critical sections become scalability burdens. In typical transaction processing systems the centralized lock manager is often the first contended component and scalability bottleneck. In this paper, we identify the conventional thread-to-transaction assignment policy as the primary cause of contention. Then, we design DORA, a system that decomposes each transaction to smaller actions and assigns actions to threads based on which data each action is about to access. DORAs design allows each thread to mostly access thread-local data structures, minimizing interaction with the contention-prone centralized lock manager. Built on top of a conventional storage engine, DORA maintains all the ACID properties. Evaluation of a prototype implementation of DORA on a multicore system demonstrates that DORA attains up to 4.8x higher throughput than a state-of-the-art storage engine when running a variety of synthetic and real-world OLTP workloads.

NoDB: efficient query execution on raw data files

As data collections become larger and larger, data loading evolves to a major bottleneck. Many applications already avoid using database systems, e.g., scientific data analysis and social networks, due to the complexity and the increased data-to-query time. For such applications data collections keep growing fast, even on a daily basis, and we are already in the era of data deluge where we have much more data than what we can move, store, let alone analyze. Our contribution in this paper is the design and roadmap of a new paradigm in database systems, called NoDB, which do not require data loading while still maintaining the whole feature set of a modern database system. In particular, we show how to make raw data files a first-class citizen, fully integrated with the query engine. Through our design and lessons learned by implementing the NoDB philosophy over a modern DBMS, we discuss the fundamental limitations as well as the strong opportunities that such a research path brings. We identify performance bottlenecks specific for in situ processing, namely the repeated parsing and tokenizing overhead and the expensive data type conversion costs. To address these problems, we introduce an adaptive indexing mechanism that maintains positional information to provide efficient access to raw data files, together with a flexible caching structure. Our implementation over PostgreSQL, called PostgresRaw, is able to avoid the loading cost completely, while matching the query performance of plain PostgreSQL and even outperforming it in many cases. We conclude that NoDB systems are feasible to design and implement over modern database architectures, bringing an unprecedented positive effect in usability and performance.

Aether: a scalable approach to logging

The shift to multi-core hardware brings new challenges to database systems, as the software parallelism determines performance. Even though database systems traditionally accommodate simultaneous requests, a multitude of synchronization barriers serialize execution. Write-ahead logging is a fundamental, omnipresent component in ARIES-style concurrency and recovery, and one of the most important yet-to-be addressed potential bottlenecks, especially in OLTP workloads making frequent small changes to data. In this paper, we identify four logging-related impediments to database system scalability. Each issue challenges different level in the software architecture: (a) the high volume of small-sized I/O requests may saturate the disk, (b) transactions hold locks while waiting for the log flush, (c) extensive context switching overwhelms the OS scheduler with threads executing log I/Os, and (d) contention appears as transactions serialize accesses to in-memory log data structures. We demonstrate these problems and address them with techniques that, when combined, comprise a holistic, scalable approach to logging. Our solution achieves a 20%-69% speedup over a modern database system when running log-intensive workloads, such as the TPC-B and TATP benchmarks. Moreover, it achieves log insert throughput over 1.8GB/s for small log records on a single socket server, an order of magnitude higher than the traditional way of accessing the log using a single mutex.The shift to multi-core hardware brings new challenges to database systems, as the software parallelism determines performance. Even though database systems traditionally accommodate simultaneous requests, a multitude of synchronization barriers serialize execution. Write-ahead logging is a fundamental, omnipresent component in ARIES-style concurrency and recovery, and one of the most important yet-to-be addressed potential bottlenecks, especially in OLTP workloads making frequent small changes to data. In this paper, we identify four logging-related impediments to database system scalability. Each issue challenges different level in the software architecture: (a) the high volume of small-sized I/O requests may saturate the disk, (b) transactions hold locks while waiting for the log flush, (c) extensive context switching overwhelms the OS scheduler with threads executing log I/Os, and (d) contention appears as transactions serialize accesses to in-memory log data structures. We demonstrate these problems and address them with techniques that, when combined, comprise a holistic, scalable approach to logging. Our solution achieves a 20%-69% speedup over a modern database system when running log-intensive workloads, such as the TPC-B and TATP benchmarks. Moreover, it achieves log insert throughput over 1.8GB/s for small log records on a single socket server, an order of magnitude higher than the traditional way of accessing the log using a single mutex.

Managing scientific data

Needed are generic, rather than one-off, DBMS solutions automating storage and analysis of data from scientific collaborations.

Optimal Service Pricing for a Cloud Cache

Cloud applications that offer data management services are emerging. Such clouds support caching of data in order to provide quality query services. The users can query the cloud data, paying the price for the infrastructure they use. Cloud management necessitates an economy that manages the service of multiple users in an efficient, but also, resource-economic way that allows for cloud profit. Naturally, the maximization of cloud profit given some guarantees for user satisfaction presumes an appropriate price-demand model that enables optimal pricing of query services. The model should be plausible in that it reflects the correlation of cache structures involved in the queries. Optimal pricing is achieved based on a dynamic pricing scheme that adapts to time changes. This paper proposes a novel price-demand model designed for a cloud cache and a dynamic pricing scheme for queries executed in the cloud cache. The pricing solution employs a novel method that estimates the correlations of the cache services in an time-efficient manner. The experimental study shows the efficiency of the solution.