Performance critical transaction and storage systems require fast persistence of write data. Typically, a non-volatile RAM (NVRAM) is employed on the datapath to the permanent storage, to temporarily and quickly store write data before the system acknowledges the write request. NVRAM is commonly implemented with battery-backed DRAM. Unfortunately, battery-backed DRAM is small and costly, and occupies a precious DIMM slot. In this paper, we make a case for dual, byte- and block-addressable solid-state drive (2B-SSD), a novel NAND flash SSD architecture designed to offer a dual view of byte addressability and traditional block addressability at the same time. Unlike a conventional storage device, 2B-SSD allows accessing the same file with two independent byte- and block-I/O paths. It controls the data transfer between its internal DRAM and NAND flash memory through an intuitive software interface, and manages the mapping of the two address spaces. 2B-SSD realizes a wholly different way and speed of accessing files on a storage device; applications can access them directly using memory-mapped I/O, and moreover write with a DRAM-like latency. To quantify the benefits of 2B-SSD, we modified logging subsystems of major database engines to store log records directly on it without buffering them in the host memory. When running popular workloads, we measured throughput gains in the range of 1.2X and 2.8X with no risk of data loss.

This work is dedicated to eliminating the overhead required for guaranteeing the storage order in the modern IO stack. The existing block device adopts a prohibitively expensive approach in ensuring the storage order among write requests: interleaving the write requests with Transfer-and-Flush . For exploiting the cache barrier command for flash storage, we overhaul the IO scheduler, the dispatch module, and the filesystem so that these layers are orchestrated to preserve the ordering condition imposed by the application with which the associated data blocks are made durable. The key ingredients of Barrier-Enabled IO stack are Epoch-based IO scheduling , Order-Preserving Dispatch , and Dual-Mode Journaling . Barrier-enabled IO stack can control the storage order without Transfer-and-Flush overhead. We implement the barrier-enabled IO stack in server as well as in mobile platforms. SQLite performance increases by 270% and 75%, in server and in smartphone, respectively. In a server storage, BarrierFS brings as much as by 43 × and by 73× performance gain in MySQL and SQLite, respectively, against EXT4 via relaxing the durability of a transaction.

Recently, in-memory big data processing frameworks have emerged, such as Apache Spark and Ignite, to accelerate workloads requiring frequent data reuse. With effective in-memory caching these frameworks eliminate most of I/O operations, which would otherwise be necessary for communication between producer and consumer tasks. However, this performance benefit is nullified if the memory footprint exceeds available memory size, due to excessive spill and garbage collection (GC) operations. To fit the working set in memory, two system parameters play an important role: number of data partitions (Npartitions) specifying task granularity, and number of tasks per each executor (Nthreads) specifying the degree of parallelism in execution. Existing approaches to optimizing these parameters either do not take into account workload characteristics, or optimize only one of the parameters in isolation, thus yielding suboptimal performance. This paper introduces WASP, a workload-aware task scheduler and partitioner, which jointly optimizes both parameters at runtime. To find an optimal setting, WASP first analyzes the DAG structure of a given workload, and uses an analytical model to predict optimal settings of Npartitionsand Nthreadsfor all stages based on their computation types. Taking this as input, the WASP scheduler employs a hill climbing algorithm to find an optimal Nthreadsfor each stage, thus maximizing concurrency while minimizing data spills and GCs. We prototype WASP on Spark and evaluate it using six workloads on three different parallel platforms. WASP improves performance by up to 3.22× and reduces the cluster operating cost on cloud by up to 40%, over the baseline following Spark Tuning Guidelines and provides robust performance for both shuffle-heavy and shuffle-light workloads.

Mobile phones security is becoming an important issue because they are being connected to an Internet through wireless modem technologies. System virtualization technology provides trusted computing capability by running isolated multiple virtual machines under hypervisor. In this paper, we propose a design of system virtualization for ARM CPU architecture and describe implementation of prototype called Xen on ARM using Xen hypervisor. Secure and nonsecure guest Linux virtual machines are executing under Xen on ARM isolated with each other and virtualization overhead is shown to be moderate compared to native Linux running on bare metal H/W.

Stateless computing is a new computing service focused on migrating user's computing contexts among different devices. In the near future, seamless computing will make computing environment easier to use and human centric. Users will not have spend time doing computing environment setup such as installation and upgrade. They will bring their own computing contexts, which consist of programs, files, and execution states, inside their pockets as saved in a small memory stick. The computing contexts will be restored on various devices, e.g. TV, desktops, mobile phone, PDA, etc. Our demonstration will show the early step to the seamless computing vision. We will demonstrate a seamless computing prototype which migrates computing contexts among different mobile embedded devices. The core technology inside the prototype which enables this computing state capability is the embedded virtual machine monitor (VMM), named as Secure Xen on ARM. We will migrate the whole state of a virtual machine between two HW platforms. Secure Xen on ARM run on the two platforms. A virtual machine running under the VMM on one platform is saved during its execution to USB memory stick, then the memory stick is plugged in to the other platform and the saved context is restored and continue execution of the virtual machine.

We will demonstrate a prototype of the beyond 3G mobile terminal based on system virtualization technology, which meets both security and flexibility requirements. In the proposed architecture, two domains, secure domain and normal domain, run on top of secure virtual machine monitor (VMM) which includes mandatory access control and domain integrity check mechanisms. Only security critical applications and services (e.g., banking, trading, and DRM) can run on the secure domain, and malware is prohibited from accessing the secure domain. Both flexibility and rich user experiences are provided by allowing users to install and execute application software, which might be potentially vulnerable, on the normal domain. Even in the case of malware attack, it cannot cause security threats to the secure domain because of the isolation and our access control features. This demonstration will show how secure our mobile terminal is against malware attacks compared to conventional single OS based mobile terminal.

A fast path planning method by optimization of a path graph for both efficiency and accuracy is proposed. A conventional quadtree-based path planning approach is simple, robust, and efficient. However, it has two limitations. We propose a path graph optimization technique employing a compact mesh representation. A world space is triangulated into a base mesh and the base mesh is simplified to a compact mesh. The compact mesh representation is object-dependent; the positions of vertexes of the mesh are optimized according to the curvatures of the obstacles. The compact mesh represents the obstacles as accurately as the quadtree even though using much fewer vertexes than the quadtree. The compact mesh distributes vertexes in a free space in a balanced way by ensuring that the lengths of edges are below an edge length threshold. An optimized path graph is extracted from the compact mesh. An iterative vertex pushing method is proposed to include important obstacle boundary edges in the path graph. Dijkstra's shortest path searching algorithm is used to search the shortest path in the path graph. Experimental results show that the path planning using the optimized path graph is an order of magnitude faster than the quadtree approach while the length of the path generated by the proposed method is almost the same as that of the path generated by the quadtree.