New-Sum: A Novel Online ABFT Scheme For General Iterative Methods
New-Sum: A Novel Online ABFT Scheme For General Iterative Methods
复制标题
New-Sum：一种新颖的通用迭代方法在线 ABFT 方案
DOI：
--
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
Zizhong Chen
中科院分区：
文献类型：
--
作者：
Dingwen Tao;S. Song;S. Krishnamoorthy;Panruo Wu;Xin Liang;E. Zhang;D. Kerbyson;Zizhong Chen
新兴的高性能计算平台具有大量组件和较低的功率裕度，预计更容易受到逻辑电路和内存子系统中软错误的影响。我们提出了一种基于在线算法的容错（ABFT）方法，可以有效地检测和恢复通用迭代方法的软错误。我们设计了一种新颖的基于校验和的矩阵向量乘法编码方案，该方案对算术错误和内存错误都有弹性。我们的设计将校验和更新过程与实际计算分离，并允许自适应校验和开销控制。基于这种新的编码机制，我们提出了两种在线 ABFT 设计，与检查点/回滚方案结合时可以有效地从错误中恢复。这些设计能够解决不同错误率下的场景。我们的 ABFT 方法适用于主要依赖于矩阵向量乘法和向量线性运算的各种迭代求解器。我们通过全面的分析和实证分析来评估我们的设计。 Stampede 超级计算机上的实验评估表明，对于 UFL 稀疏矩阵集合中最大的 SPD 矩阵，我们的两种 ABFT 方案（针对预条件 CG（0.4% 和 2.2%）和预条件 BiCGSTAB（1.0% 和 4.0%））带来的性能开销较低。该评估还证明了我们提出的设计在通用迭代方法中检测和恢复各种类型的软错误的灵活性和有效性。
Emerging high-performance computing platforms, with large component counts and lower power margins, are anticipated to be more susceptible to soft errors in both logic circuits and memory subsystems. We present an online algorithm-based fault tolerance (ABFT) approach to efficiently detect and recover soft errors for general iterative methods. We design a novel checksum-based encoding scheme for matrix-vector multiplication that is resilient to both arithmetic and memory errors. Our design decouples the checksum updating process from the actual computation, and allows adaptive checksum overhead control. Building on this new encoding mechanism, we propose two online ABFT designs that can effectively recover from errors when combined with a checkpoint/rollback scheme. These designs are capable of addressing scenarios under different error rates. Our ABFT approaches apply to a wide range of iterative solvers that primarily rely on matrix-vector multiplication and vector linear operations. We evaluate our designs through comprehensive analytical and empirical analysis. Experimental evaluation on the Stampede supercomputer demonstrates the low performance overheads incurred by our two ABFT schemes for preconditioned CG (0.4% and 2.2%) and preconditioned BiCGSTAB (1.0% and 4.0%) for the largest SPD matrix from UFL Sparse Matrix Collection. The evaluation also demonstrates the flexibility and effectiveness of our proposed designs for detecting and recovering various types of soft errors in general iterative methods.