Xie Qi is a senior architect of Intel Big Data team. He once worked for IT Flags at Intel and joined Intel Big Data team in 2016 and has a broad experience across Big Data, Multi Media and Wireless.
The capacity of data grows rapidly in big data area, more and more memory are consumed either in the computation or holding the intermediate data for analytic jobs. For those memory intensive workloads, end-point users have to scale out the computation cluster or extend memory with storage like HDD or SSD to meet the requirement of computing tasks. Intel Optane DC persistent memory (DCPM) brings higher capacity than memory and higher bandwidth & lower latency than storage like SSD or HDD. Extension storage memory with Intel Optane DCPM to cache RDDs, memory intensive workloads benefit from large memory capacity. In this session, we will talk more about how we can use DCPM to extend Spark storage memory as well as its performance advantage.
Enterprise and cloud data centers are under pressure to continuously expand revenue-generating and value-added services, such as compute intensive and I/O-demanding Big Data solutions, which moves large amounts of data into and out of storage, and sends it across the networked clusters.
A significant amount of time and network bandwidth can be saved when the data is compressed before it is passed between servers, as long as the compression/decompression operations are efficient and require negligible CPU cycles. Intel QuickAssist Technology allows compute-intensive workloads, specifically compression, to be offloaded from the CPU core onto dedicated hardware accelerators. Intel Quick Assist Technology enables developers to create software solutions that leverage compression/decompression acceleration, accessing the technology through APIs in the Intel QuickAssist Software.
This talk provides developers with information on Intel QuickAssist Technology and presents some key use cases to provide background for them to understand how they can take advantage of the hardware-based compression acceleration and performance improvements available with Intel QuickAssist Technology in their Spark applications.
Nowadays, Fieldâ-Programmable Gate Array (FPGA) is widely used on data centers, and for a wide range of data center workloads, FPGA-enabled data centers have shown greate potential for providing dramatically speed performance and energy efficiency improvement. So how to efficiently integrate FPGAs to accelerate popular frameworks for big data processing like Apache Spark is an interesting topic. In this talk, We are going to present the feasibility of incorporating FPGA acceleration into Spark based on the Intel recently-announced FPGA Programmable Acceleration Cards (PACs) for Xeon servers and using the TPCx-HS, the industry standard for benchmarking big data systems, to show that acceleration is possible. With a step-by-step case study for the TPCx-HS, we demonstrate how a straightforward integration with FPGA can offer an efficient integration with 1.2x overall system speedup and more energy efficiency improvement. Session hashtag: #SAISEco1
Nowadays, people are creating, sharing and storing data at a faster pace than ever before, effective data compression / decompression could significantly reduce the cost of data usage. Apache Spark is a general distributed computing engine for big data analytics, and it has large amount of data storing and shuffling across cluster in runtime, the data compression/decompression codecs can impact the end to end application performance in many ways. However, there's a trade-off between the storage size and compression/decompression throughput (CPU computation). Balancing the data compress speed and ratio is a very interesting topic, particularly while both software algorithms and the CPU instruction set keep evolving. Apache Spark provides a very flexible compression codecs interface with default implementations like GZip, Snappy, LZ4, ZSTD etc. and Intel Big Data Technologies team also implemented more codecs based on latest Intel platform like ISA-L(igzip), LZ4-IPP, Zlib-IPP and ZSTD for Apache Spark; in this session, we'd like to compare the characteristics of those algorithms and implementations, by running different micro workloads as well as end to end workloads, based on different generations of Intel x86 platform and disk. It's supposedly to be the best practice for big data software engineers to choose the proper compression/decompression codecs for their applications, and we also will present the methodologies of measuring and tuning the performance bottlenecks for typical Apache Spark workloads. Session hashtag: #Exp1SAIS
In this session we will present a Configurable FPGA-Based Spark SQL Acceleration Architecture. It is target to leverage FPGA highly parallel computing capability to accelerate Spark SQL Query and for FPGA's higher power efficiency than CPU we can lower the power consumption at the same time. The Architecture consists of SQL query decomposition algorithms, fine-grained FPGA based Engine Units which perform basic computation of sub string, arithmetic and logic operations. Using SQL query decomposition algorithm, we are able to decompose a complex SQL query into basic operations and according to their patterns each is fed into an Engine Unit. SQL Engine Units are highly configurable and can be chained together to perform complex Spark SQL queries, finally one SQL query is transformed into a Hardware Pipeline. We will present the performance benchmark results comparing the queries with FGPA-Based Spark SQL Acceleration Architecture on XEON E5 and FPGA to the ones with Spark SQL Query on XEON E5 with 10X ~ 100X improvement and we will demonstrate one SQL query workload from a real customer. Session hashtag: #EUres0Learn more: