Senior Member Of Technical Staff at Oracle Cloud Infrastructure(OCI)
Greater Seattle Area
8年云计算,云存储,企业级存储软件开发,运维经验。 精通亚马逊AWS平台工具,高可用容错企业级存储设计。 More than 8 years of experience in Cloud-based Storage development & operation, enterprise-storage software development, HA/large-scale fault tolerance system design, network & storage protocols. Experience Agile development, Testing, Operation,and Deployment of Web Services based on AWS platform. Experience with building cloud-based storage service on Data Center from scratch. Experience with big data infrastructure, high performance...
8年云计算,云存储,企业级存储软件开发,运维经验。 精通亚马逊AWS平台工具,高可用容错企业级存储设计。 More than 8 years of experience in Cloud-based Storage development & operation, enterprise-storage software development, HA/large-scale fault tolerance system design, network & storage protocols. Experience Agile development, Testing, Operation,and Deployment of Web Services based on AWS platform. Experience with building cloud-based storage service on Data Center from scratch. Experience with big data infrastructure, high performance Hadoop cluster, HBase, HIVE, Spark. Experience with server design, hardware bring-up and high performance computing Experience with system performance analysis, tuning and code optimization Experience with all kinds of AWS cloud tools: S3/EC2/Dynamo DB/RDS/SQS/EMR/ Proven track record of leading multiple projects and delivering them with high quality End-to-end understanding of complex software and hardware systems Experience with the entire product life cycle: design, implementation, testing, delivery and maintenance Research Interests: Service-oriented software solution. High available, auto-scaling, resilient Web Service Big data infrastructure, autoscaling/load balance design, per-user throttling, cost-control. No SQL database design High performance Hadoop Clusters. Commodity-based software defined storage. Active-active design, NVRAM/NVMEM mirroring with high performance RDMA, N-way HA. Low-latency and high bandwidth data transfer protocols over Ethernet, InfiniBand. RDMA over converged ethernet(RoCE) high available storage clusters.AWS Storage Software Engineer @ S3/Glacier: Hand-on experience on building web-based scalable hot/cold data storage from scratch. hand-on experience on large-scale data processing engine on Elastic Map-Reducing. Hand-on experience on implementing erasure coding. Hand-on experience on web-service operations, performance tuning and trouble-shooting. Billion-level manifests Reconstruction in S3. Non-disruptive on-line Job schema migration. Prevent Aging Dynamo DB table's partition design: DDB table rotation. Large-scale auto-scaling Job accept panel. Experience with NoSQL high performance/available distributed database design (DynamoDB, Cassandra, HBase and etc.). Experience with high-scalable web service application design, dealing with various tradeoffs between consistency, availability and partitions. in-memory/disk-based high available scalable key-value data store design and implementation. Hands-on Experience on varieties of design pattern, web service framework (Spring/GUICE/Guava/Lombok/) Design varieties of access Throttle/memory cache/rate limit/ failure-retry/ From July 2014 to Present (1 year 6 months) Greater Seattle AreaTechnical Staff Member @ Software Engineer in Platform Group Develop RDMA libraries for Chelsio T3 (cxgb/iwarp), mellanox CX2/CX3 (mlx4). Develop FreeBSD Kernel Device. Develop FreeBSD Networking Stack: VNET, IPSPACE, and etc. Scale-out Networking. High Availability Storage System. From June 2012 to Present (3 years 7 months) Summer Internship @ Develop iWARP driver over FreeBSD Develop a FreeBSD kernel module `testrdma' as CLI to verify ported iWARP driver From June 2012 to October 2012 (5 months) Active Contributor @ Developed a kernel-assisted MPI intra-node collective module with one-sided single-copy IPC (KNEM) to speed up large messages’ collective operations, including broadcast, gather, scatter, all-to-all and all-gather; Developed HierKNEM, a kernel-assisted and topology-aware inter- and intra-node collective component. HierKNEM combines topology-aware technique, kernel-assisted IPC(KNEM) and pipelining strategies to overlap inter- and intra-node communication to maximize throughputs of broadcast and reduction operations on multi-core clusters. On 20Gbps/40Gbps InfiniBand clusters from Grid5000, HierKNEM can beat MVAPICH2 by up to 50% in aggregate bandwidth; Developed the ELAN Byte Transfer Layer (BTL) with ELAN 3’s Tport and RDMA primitives to support Open MPI’s point-to-point communication over Quadrics networks; Developed SiCortex BTL with SiCortex DMA primitives to support Open MPI point-to-point communication over SiCortex high performance networks to achieve 0.7ns latency and 1.2GB/s bandwidth in off-node ping-pong tests; Developed SM/KNEM BTL with Kernel-Nemesis to speed up large messages on shared memory multi-core nodes, which saves CPU cycles and memory bandwidth. From July 2006 to April 2012 (5 years 10 months) Research Assistant @ Main contributor: FT-MPI FT-MPI has been developed in the frame of the HARNESS project. The goal of FT-MPI is to provide the end-user a communication library providing an MPI API, which benefits from the fault-tolerance in the HARNESS system. Therefore, FT-MPI implements the whole MPI-1.2 specification, some parts of the MPI-2 document and extends some of the semantics of MPI for giving the application the possibility to recover from failed processes. FT-MPI survives the crash of n-1 processes in a n-process job, and, if required, can respawn them. However, it is still the responsebility of the application to recover the data-structures and the data on the crahsed processes. Active Developer in DAGuE DAGuE aims at enabling scientific computing on large scale distributed environments featuring many cores, accelerators and high speed networks. The framework includes libraries, a runtime system, and development tools to help application developers tackle the difficult task of porting their applications to highly heterogeneous and diverse environment. From 2006 to 2012 (6 years) Volunteer @ working for fiber team and routing group in supporting of SuperComputing Conferences. From 2008 to 2011 (3 years) Research Internship @ Developed a non-blocking MPI-Style communication library for Roadrunner (#1 in 2009’s top 500 supercomputers), a heterogeneous InfiniBand cluster with IBM celll accelerator. IBM PowerXcell 8i one-sided rDMA is used to speed up communication inside compute nodes. From May 2008 to August 2008 (4 months) Doctor of Philosophy (Ph.D.), Computer Science, 3.8/4.0 @ The University of Tennessee From 2008 to 2012 Master of Science (M.S.), hpc, high performance interconnect, 3.79 @ University of Tennessee-Knoxville From 2006 to 2008 M.E./B.E, Computer Engineering @ Southeast University Teng Ma is skilled in: High Performance Computing, Java, Hadoop, NoSQL, dynamodb, Guice, Spring Framework, Cassandra, HBase, JUnit, clojure, go, AWS, Amazon Web Services (AWS), TestNG, Mockito, Spark, EMR
Amazon
AWS Storage Software Engineer
July 2014 to Present
Greater Seattle Area
NetApp
Technical Staff Member
June 2012 to Present
NetApp
Summer Internship
June 2012 to October 2012
Open MPI
Active Contributor
July 2006 to April 2012
Innovative Computing Laboratory (ICL)
Research Assistant
2006 to 2012
SciNet High Performance Computing Consortium
Volunteer
2008 to 2011
Los Alamos National Laboratory
Research Internship
May 2008 to August 2008
S3/Glacier: Hand-on experience on building web-based scalable hot/cold data storage from scratch. hand-on experience on large-scale data processing engine on Elastic Map-Reducing. Hand-on experience on implementing erasure coding. Hand-on experience on web-service operations, performance tuning and trouble-shooting. Billion-level manifests Reconstruction in S3. Non-disruptive on-line Job schema migration. Prevent Aging Dynamo DB table's partition design: DDB table rotation.... S3/Glacier: Hand-on experience on building web-based scalable hot/cold data storage from scratch. hand-on experience on large-scale data processing engine on Elastic Map-Reducing. Hand-on experience on implementing erasure coding. Hand-on experience on web-service operations, performance tuning and trouble-shooting. Billion-level manifests Reconstruction in S3. Non-disruptive on-line Job schema migration. Prevent Aging Dynamo DB table's partition design: DDB table rotation. Large-scale auto-scaling Job accept panel. Experience with NoSQL high performance/available distributed database design (DynamoDB, Cassandra, HBase and etc.). Experience with high-scalable web service application design, dealing with various tradeoffs between consistency, availability and partitions. in-memory/disk-based high available scalable key-value data store design and implementation. Hands-on Experience on varieties of design pattern, web service framework (Spring/GUICE/Guava/Lombok/) Design varieties of access Throttle/memory cache/rate limit/ failure-retry/
What company does Teng Ma work for?
Teng Ma works for Amazon
What is Teng Ma's role at Amazon?
Teng Ma is AWS Storage Software Engineer
What industry does Teng Ma work in?
Teng Ma works in the Computer Networking industry.
Who are Teng Ma's colleagues?
Teng Ma's colleagues are Apurv Mahajan, Cheng-Hao Kuo, Ezhilan Ilangovan, Brian Tanner, Peter Popp, Sarah Hammond, Mahesh Lunawat, Shrikant Thakare, Matthew Johnston, and Clint Solomon
Enjoy unlimited access and discover candidates outside of LinkedIn
One billion email addresses and counting
Everything you need to engage with more prospects.
ContactOut is used by
76% of Fortune 500 companies