Busy-wait techniques are heavily used for mutual exclusion and barrier synchronization in shared-memory parallel programs. Unfortunately, typical implementations of busy-waiting tend to produce large amounts of memory and interconnect contention, introducing performance bottlenecks that become markedly more pronounced as applications scale. We argue that this problem is not fundamental, and that one can in fact construct busy-wait synchronization algorithms that induce no memory or interconnect contention. The key to these algorithms is for every processor to spin on separate locally-accessible flag variables, and for some other processor to terminate the spin with a single remote write operation at an appropriate time. Flag variables may be locally-accessible as a result of coherent caching, or by virtue of allocation in the local portion of physically distributed shared memory.We present a new scalable algorithm for spin locks that generates 0(1) remote references per lock acquisition, independent of the number of processors attempting to acquire the lock. Our algorithm provides reasonable latency in the absence of contention, requires only a constant amount of space per lock, and requires no hardware support other than a swap-with-memory instruction. We also present a new scalable barrier algorithm that generates 0(1) remote references per processor reaching the barrier, and observe that two previously-known barriers can likewise be cast in a form that spins only on locally-accessible flag variables. None of these barrier algorithms requires hardware support beyond the usual atomicity of memory reads and writes.We compare the performance of our scalable algorithms with other software approaches to busy-wait synchronization on both a Sequent Symmetry and a BBN Butterfly. Our principal conclusion is that contention due to synchronization need not be a problem in large-scale shared-memory multiprocessors. The existence of scalable algorithms greatly weakens the case for costly special-purpose hardware support for synchronization, and provides a case against so-called “dance hall” architectures, in which shared memory locations are equally far from all processors. —From the Authors' Abstract

/pdf/algorithms-for-scalable-synchronization-on-shared-memory-2fqah2bkze.pdf

Algorithms for scalable synchronization on shared-memory multiprocessors

13. 벼잎선충의 약제처리 방법별 방제 효과

Recent technological advances in the development of flash-memory based devices have consolidated their leadership position as the preferred storage media in the embedded systems market and opened new vistas for deployment in enterprise-scale storage systems. Unlike hard disks, flash devices are free from any mechanical moving parts, have no seek or rotational delays and consume lower power. However, the internal idiosyncrasies of flash technology make its performance highly dependent on workload characteristics. The poor performance of random writes has been a cause of major concern, which needs to be addressed to better utilize the potential of flash in enterprise-scale environments. We examine one of the important causes of this poor performance: the design of the Flash Translation Layer (FTL), which performs the virtual-to-physical address translations and hides the erase-before-write characteristics of flash. We propose a complete paradigm shift in the design of the core FTL engine from the existing techniques with our Demand-based Flash Translation Layer (DFTL), which selectively caches page-level address mappings. We develop a flash simulation framework called FlashSim. Our experimental evaluation with realistic enterprise-scale workloads endorses the utility of DFTL in enterprise-scale storage systems by demonstrating: (i) improved performance, (ii) reduced garbage collection overhead and (iii) better overload behavior compared to state-of-the-art FTL schemes. For example, a predominantly random-write dominant I/O trace from an OLTP application running at a large financial institution shows a 78% improvement in average response time (due to a 3-fold reduction in operations of the garbage collector), compared to a state-of-the-art FTL scheme. Even for the well-known read-dominant TPC-H benchmark, for which DFTL introduces additional overheads, we improve system response time by 56%.

/pdf/dftl-a-flash-translation-layer-employing-demand-based-1jqzm0wlg2.pdf

DFTL: a flash translation layer employing demand-based selective caching of page-level address mappings

Replicating data under Eventual Consistency (EC) allows any replica to accept updates without remote synchronisation. This ensures performance and scalability in large-scale distributed systems (e.g., clouds). However, published EC approaches are ad-hoc and error-prone. Under a formal Strong Eventual Consistency (SEC) model, we study sufficient conditions for convergence. A data type that satisfies these conditions is called a Conflict-free Replicated Data Type (CRDT). Replicas of any CRDT are guaranteed to converge in a self-stabilising manner, despite any number of failures. This paper formalises two popular approaches (state- and operation-based) and their relevant sufficient conditions. We study a number of useful CRDTs, such as sets with clean semantics, supporting both add and remove operations, and consider in depth the more complex Graph data type. CRDT types can be composed to develop large-scale distributed applications, and have interesting theoretical properties.

Conflict-free Replicated Data Types

In NAND flash-based storage systems, an intermediate software layer called a flash translation layer (FTL)is usually employed to hide the erase-before-write characteristics of NAND flash memory. This paper proposes a novel superblockbased FTL scheme, which combines a set of adjacent logical blocks into a superblock. In the proposed FTL scheme, superblocks are mapped at coarse granularity,while pages inside the superblock are mapped freely at fine granularity to any location in several physical blocks. To reduce extra storage and flash memory operations, the fine-grain mapping information is stored in the spare area of NAND flash memory. This hybrid mapping technique has the flexibility provided by fine-grain address translation, while reducing the memory overhead to the level of coarse-grain address translation. Our experimental results show that the proposed FTL scheme decreases the garbage collection overhead up to 40% compared to previous FTL schemes.

/pdf/a-superblock-based-flash-translation-layer-for-nand-flash-46orzb4ip2.pdf

A superblock-based flash translation layer for NAND flash memory

In most operating systems which are customized for disk-based storage system, the replacement algorithm concerns only the number of memory hits. However, flash memory has different read and write cost in the aspects of time and energy so the replacement algorithm with flash memory should consider not only the hit count but also the replacement cost caused by selecting dirty victims. The replacement cost of dirty page is higher than that of clean page with regard to both access time and energy consumption. In this paper, we propose the Clean-First LRU (CFLRU) replacement algorithm that exploits the characteristics of flash memory. CFLRU splits the LRU list into the working region and the clean-first region and adopts a policy that evicts clean pages preferentially in the clean-first region until the number of page hits in the working region is preserved in a suitable level. Using the trace-driven simulation, the proposed algorithm reduces the average replacement cost by 28.4% in swap system and by 26.2% in buffer cache, compared with LRU algorithm. We also implement the CFLRU algorithm in the Linux kernel and present some optimization issues.

http://csl.skku.edu/papers/CS-TR-2006-250.pdf

CFLRU: a replacement algorithm for flash memory

This paper presents a novel buffer management scheme for portable media players equipped with flash memory. Though flash memory has various advantages over magnetic disks such as small and lightweight form factor, solid-state reliability, low power consumption, and shock resistance, its physical characteristics imposes several limitations. Most notably, it takes relatively long time to write data in flash memory and the data cannot be overwritten before being erased first. Since an erase operation is performed as a unit of larger block, the employed strategy for mapping logical blocks onto physical pages affects real performance of flash memory. This article suggests a flash-aware buffer management scheme that reduces the number of erase operations by selecting a victim based on its page utilization rather than based on the traditional LRU policy. Our scheme effectively minimizes the number of write and erase operations in flash memory, reducing the total execution time by 17% compared to the LRU policy.

/pdf/fab-flash-aware-buffer-management-policy-for-portable-media-2k6bm6jpfj.pdf

FAB: flash-aware buffer management policy for portable media players

A multi-channel architecture for high-performance NAND flash-based storage system

The use of virtualization is progressively accommodating diverse and unpredictable workloads as being adopted in virtual desktop and cloud computing environments. Since a virtual machine monitor lacks knowledge of each virtual machine, the unpredictableness of workloads makes resource allocation difficult. Particularly, virtual machine scheduling has a critical impact on I/O performance in cases where the virtual machine monitor is agnostic about the internal workloads of virtual machines. This paper presents a task-aware virtual machine scheduling mechanism based on inference techniques using gray-box knowledge. The proposed mechanism infers the I/O-boundness of guest-level tasks and correlates incoming events with I/O-bound tasks. With this information, we introduce partial boosting, which is a priority boosting mechanism with task-level granularity, so that an I/O-bound task is selectively scheduled to handle its incoming events promptly. Our technique focuses on improving the performance of I/O-bound tasks within heterogeneous workloads by lightweight mechanisms with complete CPU fairness among virtual machines. All implementation is confined to the virtualization layer based on the Xen virtual machine monitor and the credit scheduler. We evaluate our prototype in terms of I/O performance and CPU fairness over synthetic mixed workloads and realistic applications.

/pdf/task-aware-virtual-machine-scheduling-for-i-o-performance-46rthpkp3d.pdf

Joonwon Lee

Papers

A superblock-based flash translation layer for NAND flash memory

CFLRU: a replacement algorithm for flash memory

FAB: flash-aware buffer management policy for portable media players

A multi-channel architecture for high-performance NAND flash-based storage system

Task-aware virtual machine scheduling for I/O performance.