As we move towards hyper-converged cloud solutions, the efficiency and overheads of distributed file systems at the cloud tenant side (i.e., client) become of paramount importance. Often, the clientside driver of a cloud file system is complex and CPU intensive, deeply coupled with the backend implementation, and requires optimizing multiple intrusive knobs. In this work, we propose to decouple the file system client from its backend implementation by virtualizing it with an off-the-shelf DPU using the Linux virtio-fs software stack. The decoupling allows us to offload the file system client execution to a DPU, which is managed and optimized by the cloud provider, while freeing the host CPU cycles. DPFS, our proposed framework, is 4.4× more host CPU efficient per I/O, delivers comparable performance to a tenant with zero-configuration and without modification of their host software stack, while allowing workload and hardware specific backend optimizations. The DPFS framework and its artifacts are publically available at https://github.com/IBM/DPFS.

The continuous growth in 3D-NAND flash storage density has primarily been enabled by 3D stacking and by increasing the number of bits stored per memory cell. Unfortunately, these desirable flash device design choices are adversely affecting reliability and latency characteristics. In particular, increasing the number of bits stored per cell results in having to apply additional voltage thresholds during each read operation, therefore increasing the read latency characteristics. While most NAND flash challenges can be mitigated through appropriate background processing, the flash read latency characteristics cannot be hidden and remains the biggest challenge, especially for the newest flash generations that store four bits per cell. In this paper, we introduce read heat separation (RHS), a new heat-aware data-placement technique that exploits the skew present in real-world workloads to place frequently read user data on low-latency flash pages. Although conceptually simple, such a technique is difficult to integrate in a flash controller, as it introduces a significant amount of complexity, requires more metadata, and is further constrained by other flash-specific peculiarities. To overcome these challenges, we propose a novel flash controller architecture supporting read heat-aware data placement. We first discuss the trade-offs that such a new design entails and analyze the key aspects that influence the efficiency of RHS. Through both, extensive simulations and an implementation we realized in a commercial enterprise-grade solid-state drive controller, we show that our architecture can indeed significantly reduce the average read latency. For certain workloads, it can reverse the system-level read latency trends when using recent multi-bit flash generations and hence outperform SSDs using previous faster flash generations.

Solid-state drives (SSDs) are extensively used to deploy persistent data stores, as they provide low latency random access, high write throughput, high data density, and low cost. Tree-based data structures are widely used to build persistent data stores, and indeed they lie at the backbone of many of the data management systems used in production and research today. We show that benchmarking a persistent tree-based data structure on an SSD is a complex process, which may easily incur subtle pitfalls that can lead to an inaccurate performance assessment. At a high-level, these pitfalls stem from the interaction of complex software running on complex hardware. On the one hand, tree structures implement internal operations that have non-trivial effects on performance. On the other hand, SSDs employ firmware logic to deal with the idiosyncrasies of the underlying flash memory, which are well known to also lead to complex performance dynamics. We identify seven benchmarking pitfalls using RocksDB and WiredTiger, two widespread implementations of an LSM-Tree and a B+Tree, respectively. We show that such pitfalls can lead to incorrect measurements of key performance indicators, hinder the reproducibility and the representativeness of the results, and lead to suboptimal deployments in production environments. We also provide guidelines on how to avoid these pitfalls to obtain more reliable performance measurements, and to perform more thorough and fair comparisons among different design points.

Over the last few years, NAND flash manufacturers have steadily increased the number of bits stored per cell to achieve significant cost reductions. However, the increased density does not come without drawbacks. All key flash performance metrics, including latency and endurance, significantly degrade as bit density increases. Particularly, sustained write throughput is the worst affected as writes are roughly one order of magnitude slower than reads and further require precursory block erases in the background. As a result, many recent flash controllers operate flash blocks both in single-bit (high endurance and performance) and in multi-bit (high density) mode. In theory, such hybrid controllers are a great way of hiding flash technology limitations. A controller can use a small percentage of the flash blocks in single-bit mode as a cache which allows orders of magnitude higher write bandwidth and endurance in environments where the access patterns of the workload are skewed and bursty. In practice, however, many devices fall short of expectations when write performance varies significantly and utilization increases. We argue that a principled approach is required to understand the design trade-offs of hybrid NAND flash controllers. To this end, we develop a modeling framework for estimating the performance and endurance of hybrid controllers. The modeling framework computes the internal data movement generated by a hybrid controller by relying on advanced analytical models that offer both accurate and fast predictions. The data flow is then translated into higher-level metrics that quantify upper bounds for the overall performance of an SSD such as write throughput, latency, and device endurance. Using our modeling framework, we compare different controller architectures, identify their strong and weak points, and show that there is room to improve the efficiency of the hybrid controllers used today.

The database research community has been building methods to store, access, and update data for more than four decades. Throughout the evolution of the structures and techniques used to access data, access methods adapt to the ever changing hardware and workload requirements. Today, even small changes in the workload or the hardware lead to a redesign of access methods. The need for new designs has been increasing as data generation and workload diversification grow exponentially, and hardware advances introduce increased complexity. New workload requirements are introduced by the emergence of new applications, and data is managed by large systems composed of more and more complex and heterogeneous hardware. As a result, it is increasingly important to develop application-aware and hardware-aware access methods. The fundamental challenges that every researcher, systems architect, or designer faces when designing a new access method are how to minimize, i) read times (R), ii) update cost (U), and iii) memory (or storage) overhead (M). In this paper, we conjecture that when optimizing the read-update-memory overheads, optimizing in any two areas negatively impacts the third. We present a simple model of the RUM overheads, and we articulate the RUM Conjecture. We show how the RUM Conjecture manifests in stateof-the-art access methods, and we envision a trend toward RUMaware access methods for future data systems.

Data warehouses have been traditionally optimized for read-only query performance, allowing only offline updates at night, essentially trading off data freshness for performance. The need for 24x7 operations in global markets and the rise of online and other quickly reacting businesses make concurrent online updates increasingly desirable. Unfortunately, state-of-the-art approaches fall short of supporting fast analysis queries over fresh data. The conventional approach of performing updates in place can dramatically slow down query performance, while prior proposals using differential updates either require large in-memory buffers or may incur significant update migration cost. This article presents a novel approach for supporting online updates in data warehouses that overcomes the limitations of prior approaches by making judicious use of available SSDs to cache incoming updates. We model the problem of query processing with differential updates as a type of outer join between the data residing on disks and the updates residing on SSDs. We present MaSM algorithms for performing such joins and periodic migrations, with small memory footprints, low query overhead, low SSD writes, efficient in-place migration of updates, and correct ACID support. We present detailed modeling of the proposed approach, and provide proofs regarding the fundamental properties of the MaSM algorithms. Our experimentation shows that MaSM incurs only up to 7% overhead both on synthetic range scans (varying range size from 4KB to 100GB) and in a TPC-H query replay study, while also increasing the update throughput by orders of magnitude.

Even though main memory is becoming large enough to fit most OLTP databases, it may not always be the best option. OLTP workloads typically exhibit skewed access patterns where some records are hot (frequently accessed) but many records are cold (infrequently or never accessed). Therefore, it is more economical to store the coldest records on a fast secondary storage device such as a solid-state disk. However, main-memory DBMS have no knowledge of secondary storage, while traditional disk-based databases, designed for workloads where data resides on HDD, introduce too much overhead for the common case where the working set is memory resident.

Main memories are becoming sufficiently large that most OLTP databases can be stored entirely in main memory, but this may not be the best solution. OLTP workloads typically exhibit skewed access patterns where some records are hot (frequently accessed) but many records are cold (infrequently or never accessed). It is more economical to store the coldest records on secondary storage such as flash. As a first step towards managing cold data in databases optimized for main memory we investigate how to efficiently identify hot and cold data. We propose to log record accesses - possibly only a sample to reduce overhead - and perform offline analysis to estimate record access frequencies. We present four estimation algorithms based on exponential smoothing and experimentally evaluate their efficiency and accuracy. We find that exponential smoothing produces very accurate estimates, leading to higher hit rates than the best caching techniques. Our most efficient algorithm is able to analyze a log of 1B accesses in sub-second time on a workstation-class machine.

Solid-state drives (SSDs) are quickly becoming the default storage medium as the cost of NAND flash memory continues to drop. However, flash memory introduces new challenges, as data cannot be eciently updated in-place. To overcome the technology's limitations, SSDs incorporate a software Flash Translation Layer (FTL) that implements out-of-place updates, typically by storing data in a log-structured fashion. Despite a large number of existing FTL algorithms, SSD performance, predictability, and lifetime remain an issue, especially for the write-intensive workloads specific to database applications. In this paper, we show how to design FTLs that are more efficient by using the I/O write skew to guide data placement on flash memory. We model the relationship between data placement and write performance for basic I/O write patterns and detail the most important concepts of writing to flash memory: i) the trade-o between the extra capacity available and write overhead, ii) the benefit of adapting data placement to write skew, iii) the impact of the cleaning policy, and iv) how to estimate the best achievable write performance for a given I/O workload. Based on the findings of the theoretical model, we propose a new principled data placement algorithm that can be incorporated into existing FTLs. We show the benefits of our data placement algorithm when running micro-benchmarks and real database I/O traces: our data placement algorithm reduces write overhead by 20% - 75% when compared to state-of-art techniques.

Data warehouses have been traditionally optimized for read-only query performance, allowing only offline updates at night, essentially trading off data freshness for performance. The need for 24x7 operations in global markets and the rise of online and other quickly-reacting businesses make concurrent online updates increasingly desirable. Unfortunately, state-of-the-art approaches fall short of supporting fast analysis queries over fresh data. The conventional approach of performing updates in place can dramatically slow down query performance, while prior proposals using differential updates either require large in-memory buffers or may incur significant update migration cost.

Many parallel applications exhibit unpredictable communication between threads, leading to contention for shared objects. The choice of contention management strategy impacts strongly the performance and scalability of these applications: spinning provides maximum performance but wastes significant processor resources, while blocking-based approaches conserve processor resources but introduce high overheads on the critical path of computation. Under situations of high or changing load, the operating system complicates matters further with arbitrary scheduling decisions which often preempt lock holders, leading to long serialization delays until the preempted thread resumes execution.

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.

Database engines face growing scalability challenges as core counts exponentially increase each processor generation, and the efficiency of synchronization primitives used to protect internal data structures is a crucial factor in overall database performance. The trade-offs between different implementation approaches for these primitives shift significantly with increasing degrees of available hardware parallelism. Blocking synchronization, which has long been the favored approach in database systems, becomes increasingly unattractive as growing core counts expose its bottlenecks. Spinning implementations improve peak system throughput by a factor of 2x or more for 64 hardware contexts, but suffer from poor performance under load.

In the last few years NAND flash storage has become more and more popular as price per GB and capacity both improve at exponential rates. Flash memory offers significant benefits compared to magnetic hard disk drives (HDDs) and DBMSs are highly likely to use flash as a general storage backend, either alone or in heterogeneous storage solutions with HDDs. Flash devices, however, respond quite differently than HDDs for common access patterns, and recent research shows a strong asymmetry between read and write performance. Moreover, flash storage devices behave unpredictably, showing a high dependence on previous I/O history and usage patterns. In this paper we investigate how a DBMS can overcome these issues to take full advantage of flash memory as persistent storage. We propose new a flash aware data layout - append and pack - which stabilizes device performance by eliminating random writes. We assess the impact of append and pack on OLTP workload performance using both an analytical model and micro-benchmarks, and our results suggest that significant improvements can be achieved for real workloads.

We investigate two Markov object processes to model line networks. The first one is based on region interactions whereas the interactions of the second one are defined w.r.t. the segment intersection points. We use a Metropolis-Hastings-Green scheme to simulate both models.

This paper presents a method for indexing and retrieval of multimedia data through texture segmentation, using the Wold decomposition. The texture field is assumed to be a realisation of a regular homogeneous random field. On the basis of a 2-D Wold-like decomposition, the field is represented as the sum of a purely indeterministic component, a harmonic component and a countable number of evanescent fields. A new rigorous distance measure between textures is derived, using Wold parameters. Adopting the MRF framework, we construct a segmentation procedure using the Wold parameters.

This paper presents a method for indexing and retrieval of multimedia data. The proposed indexing and retrieval strategy is based on the usage of textural information contained in the data imagery components as the indexing keys. On the basis of a 2-D Wold-like decomposition, the texture field is represented as the sum of purely indeterministic, harmonic, and evanescent fields. A new rigorous distance measure between textures which employs their estimated parametric models, is developed. This distance measure is then applied to retrieve multimedia records that contain images with textured segments which are similar to those in a given image. Evaluation of this distance measure is computationally efficient, and hence highly suitable for data base retrieval applications.