Benjamin Eckart

BPAC: An adaptive write buffer management scheme for flash-based Solid State Drives

Solid State Drives (SSDs) have shown promise to be a candidate to replace traditional hard disk drives, but due to certain physical characteristics of NAND flash, there are some challenging areas of improvement and further research. We focus on the layout and management of the small amount of RAM that serves as a cache between the SSD and the system that uses it. Of the techniques that have previously been proposed to manage this cache, we identify several sources of inefficient cache space management due to the way pages are clustered in blocks and the limited replacement policy. We develop a hybrid page/block architecture along with an advanced replacement policy, called BPAC, or Block-Page Adaptive Cache, to exploit both temporal and spatial locality. Our technique involves adaptively partitioning the SSD on-disk cache to separately hold pages with high temporal locality in a page list and clusters of pages with low temporal but high spatial locality in a block list. We run trace-driven simulations to verify our design and find that it outperforms other popular flash-aware cache schemes under different workloads.

An adaptive write buffer management scheme for flash-based SSDs

Solid State Drives (SSDs) have shown promise to be a candidate to replace traditional hard disk drives. The benefits of SSDs over HDDs include better durability, higher performance, and lower power consumption, but due to certain physical characteristics of NAND flash, which comprise SSDs, there are some challenging areas of improvement and further research. We focus on the layout and management of the small amount of RAM that serves as a cache between the SSD and the system that uses it. Of the techniques that have previously been proposed to manage this cache, we identify several sources of inefficient cache space management due to the way pages are clustered in blocks and the limited replacement policy. We find that in many traces hot pages reside in otherwise cold blocks, and that the spatial locality of most clusters can be fully exploited in a limited time period, so we develop a hybrid page/block architecture along with an advanced replacement policy, called BPAC, or Block-Page Adaptive Cache, to exploit both temporal and spatial locality. Our technique involves adaptively partitioning the SSD on-disk cache to separately hold pages with high temporal locality in a page list and clusters of pages with low temporal but high spatial locality in a block list. In addition, we have developed a novel mechanism for flash-based SSDs to characterize the spatial locality of the disk I/O workload and an approach to dynamically identify the set of low spatial locality clusters. We run trace-driven simulations to verify our design and find that it outperforms other popular flash-aware cache schemes under different workloads. For instance, compared to a popular flash aware cache algorithm BPLRU, BPAC reduces the number of cache evictions by up to 79.6% and 34% on average.

Code-M: A non-MDS erasure code scheme to support fast recovery from up to two-disk failures in storage systems

In this paper, we present a novel coding scheme that can tolerate up to two-disk failures, satisfying the RAID-6 property. Our coding scheme, Code-M, is a non-MDS (Maximum Distance Separable, tolerating maximum failures with a given amount of redundancy) code that is optimized by trading rate for fast recovery times. Code-M is lowest density and its parity chain length is fixed at 2C − 1 for a given number of columns in a strip-set C. The rate of Code-M, or percentage of disk space occupied by non-parity data, is (C − 1)/C. We perform theoretical analysis and evaluation of the coding scheme under different configurations. Our theoretical analysis shows that Code-M has favorable reconstruction times compared to RDP, another well-established RAID-6 code. The quantitative comparisons of Code-M against RDP demonstrate recovery performance improvement by a factor of up to 5.18 under single disk failure and 2.8 under double failures using the same number of disks. Overall, Code-M is a RAID-6 type code supporting fast recovery with reduced I/O complexity.

A Dynamic Performance-Based Flow Control Method for High-Speed Data Transfer

New types of specialized network applications are being created that need to be able to transmit large amounts of data across dedicated network links. TCP fails to be a suitable method of bulk data transfer in many of these applications, giving rise to new classes of protocols designed to circumvent TCPs shortcomings. It is typical in these high-performance applications, however, that the system hardware is simply incapable of saturating the bandwidths supported by the network infrastructure. When the bottleneck for data transfer occurs in the system itself and not in the network, it is critical that the protocol scales gracefully to prevent buffer overflow and packet loss. It is therefore necessary to build a high-speed protocol adaptive to the performance of each system by including a dynamic performance-based flow control. This paper develops such a protocol, performance adaptive UDP (henceforth PA-UDP), which aims to dynamically and autonomously maximize performance under different systems. A mathematical model and related algorithms are proposed to describe the theoretical basis behind effective buffer and CPU management. A novel delay-based rate-throttling model is also demonstrated to be very accurate under diverse system latencies. Based on these models, we implemented a prototype under Linux, and the experimental results demonstrate that PA-UDP outperforms other existing high-speed protocols on commodity hardware in terms of throughput, packet loss, and CPU utilization. PA-UDP is efficient not only for high-speed research networks, but also for reliable high-performance bulk data transfer over dedicated local area networks where congestion and fairness are typically not a concern.

Performance adaptive UDP for high-speed bulk data transfer over dedicated links

New types of networks are emerging for the purpose of transmitting large amounts of scientific data among research institutions quickly and reliably. These exotic networks are characterized by being high-bandwidth, high- latency, and free from congestion. In this environment, TCP ceases to be an appropriate protocol for reliable bulk data transfer because it fails to saturate link throughput. Of the new protocols designed to take advantage of these networks, a subclass has emerged using UDP for data transfer and TCP for control. These high-speed variants of reliable UDP, however, tend to underperform on all but high-end systems due to constraints of the CPU, network, and hard disk. It is therefore necessary to build a high-speed protocol adaptive to the performance of each system. This paper develops such a protocol, Performance Adaptive UDP (henceforth PA-UDP), which aims to dynamically and autonomously maximize performance under different systems. A mathematical model and related algorithms are proposed to describe the theoretical basis behind effective buffer and CPU management. Based on this model, we implemented a prototype under Linux and the experimental results demonstrate that PA-UDP outperforms an existing high-speed protocol on commodity hardware in terms of throughput and packet loss. PA- UDP is efficient not only for high-speed research networks but also for reliable high-performance bulk data transfer over dedicated local area networks where congestion and fairness are typically not a concern.

MLMD: Maximum Likelihood Mixture Decoupling for Fast and Accurate Point Cloud Registration

Registration of Point Cloud Data (PCD) forms a core component of many 3D vision algorithms such as object matching and environment reconstruction. In this paper, we introduce a PCD registration algorithm that utilizes Gaussian Mixture Models (GMM) and a novel dual-mode parameter optimization technique which we call mixture decoupling. We show how this decoupling technique facilitates both faster and more robust registration by first optimizing over the mixture parameters (decoupling the mixture weights, means, and co variances from the points) before optimizing over the 6 DOF registration parameters. Furthermore, we frame both the decoupling and registration process inside a unified, dual-mode Expectation Maximization (EM) framework, for which we derive a Maximum Likelihood Estimation (MLE) solution along with a parallel implementation on the GPU. We evaluate our MLE-based mixture decoupling (MLMD) registration method over both synthetic and real data, showing better convergence for a wider range of initial conditions and higher speeds than previous state of the art methods.

REM-Seg: A robust EM algorithm for parallel segmentation and registration of point clouds

For purposes of real-time 3D sensing, it is important to be able to quickly register together incoming point cloud data. In this paper, we devise a method to quickly and robustly decompose large point clouds into a relatively small number of meaningful surface patches from which we register new data points. The surface patch representation sidesteps the costly problem of matching points to points since incoming data only need to be compared with the patches. The chosen parametrization of the patches (as Gaussians) leads to a smooth data likelihood function with a well-defined gradient. This representation thus forms the basis for a robust and efficient registration algorithm using a parallelized gradient descent implemented on a GPU using CUDA. We use a modified Gaussian Mixture Model (GMM) formulation solved by Expectation Maximization (EM) to segment the point cloud and an annealing gradient descent method to find the 6-DOF rigid transformation between the incoming point cloud and the segmented set of surface patches. We test our algorithm, Robust EM Segmentation (REM-Seg), against other GPU-accelerated registration algorithms on simulated and real data and show that our method scales well to large numbers of points, has a wide range of convergence, and is suitably accurate for 3D registration.

Accelerated Generative Models for 3D Point Cloud Data

Finding meaningful, structured representations of 3D point cloud data (PCD) has become a core task for spatial perception applications. In this paper we introduce a method for constructing compact generative representations of PCD at multiple levels of detail. As opposed to deterministic structures such as voxel grids or octrees, we propose probabilistic subdivisions of the data through local mixture modeling, and show how these subdivisions can provide a maximum likelihood segmentation of the data. The final representation is hierarchical, compact, parametric, and statistically derived, facilitating run-time occupancy calculations through stochastic sampling. Unlike traditional deterministic spatial subdivision methods, our technique enables dynamic creation of voxel grids according the applications best needs. In contrast to other generative models for PCD, we explicitly enforce sparsity among points and mixtures, a technique which we call expectation sparsification. This leads to a highly parallel hierarchical Expectation Maximization (EM) algorithm well-suited for the GPU and real-time execution. We explore the trade-offs between model fidelity and model size at various levels of detail, our tests showing favorable performance when compared to octree and NDT-based methods.

Distributed Virtual Diskless Checkpointing: A Highly Fault Tolerant Scheme for Virtualized Clusters

Todays high-end computing systems are facing a crisis of high failure rates due to increased numbers of components. Recent studies have shown that traditional fault tolerant techniques incur overheads that more than double execution times on these highly parallel machines. Thus, future high-end computing must be able to provide adequate fault tolerance at an acceptable cost or the burdens of fault management will severely affect the viability of such systems. Cluster virtualization offers a potentially unique solution for fault management, but brings significant overhead, especially for I/O. In this paper, we propose a novel diskless check pointing technique on clusters of virtual machines. Our technique splits Virtual Machines into sets of orthogonal RAID systems and distributes parity evenly across the cluster, similar to a RAID-5 configuration, but using VM images as data elements. Our theoretical analysis shows that our technique significantly reduces the overhead associated with check pointing by removing the disk I/O bottleneck.

An Extensible I/O Performance Analysis Framework for Distributed Environments

As distributed systems increase in both popularity and scale, it becomes increasingly important to understand as well as to systematically identify performance anomalies and potential opportunities for optimization. However, large scale distributed systems are often complex and non-deterministic due to hardware and software heterogeneity and configurable runtime options that may boost or diminish performance. It is therefore important to be able to disseminate and present the information gleaned from a local system under a common evaluation methodology so that such efforts can be valuable in one environment and provide general guidelines for other environments. Evaluation methodologies can conveniently be encapsulated inside of a common analysis framework that serves as an outer layer upon which appropriate experimental design and relevant workloads (benchmarking and profiling applications) can be supported. In this paper we present ExPerT, an Ex tensible Per formance T oolkit. ExPerT defines a flexible framework from which a set of benchmarking, tracing, and profiling applications can be correlated together in a unified interface. The framework consists primarily of two parts: an extensible module for profiling and benchmarking support, and a unified data discovery tool for information gathering and parsing. We include a case study of disk I/O performance in virtualized distributed environments which demonstrates the flexibility of our framework for selecting benchmark suite, creating experimental design, and performing analysis.