ISSC431 Benefits Parallel Processing Discussion Response | savvyessaywriters.us
Hello, I need to respond to the below two students discussions with a minimum of 150 words each. Below in bold is what the students are responding to. Thanks in advance.
- For this assignment, answer the following:
- What type of queries benefit from parallel processing?
- What are the benefits of parallel databases?
- How does Oracle utilize memory to speed up processes?
Parallel processing is simply when a large task is divided into numerous smaller tasks and the smaller tasks are performed simultaneously on multiple nodes. The performance of many types of queries is improved by utilizing parallel processing. It improves response time and enhances throughput. Parallel processing speeds up DSS queries, parallel queries, and is possible to speed up batch (mixed) queries. It allows all the previously mentioned queries along with OLTP to attain scaleup when parallel processing is correctly executed.
A parallel database offers many benefits such as higher performance, higher availability, greater flexibility, and more users. When more CPUs are used in an application, a higher speedup and scaleup becomes possible. Since nodes are separated from each other, if one fails the entire system does not go down. In an Oracle Parallel Server environment, instances can be flexible as they can be allocated or deallocated if needed. Using parallel database technology, it is feasible to disregard memory limits and a single system can have thousands of users.
Oracle utilizes several memory optimizations that help speed up database processes. The first is Non-Uniform Memory Access (NUMA) Optimization and Memory Placement. In NUMA machines, it has several processors and each processor manages memory along with the memory coherence between processors. Another optimization is large memory pages. The Oracle database can allocate up to 2 GB pages to the shared memory areas purposed for the database. One more optimization is multithreaded shared memory operations. The latest version of Oracle provides a multithreaded kernel process called vmtasks that speeds up the creation, locking, and destruction of pages stored in the shared memory.
Parallel Processing & Parallel Databases. (n.d.). Oracle8 Parallel Server Concepts & Administration, 8. Retrieved from https://docs.oracle.com/cd/A58617_01/server.804/a58238/ch1_unde.htm#3126.
Henningsen, G. (2013). How Oracle Solaris Makes Oracle Database Fast. Oracle. Retrieved from https://www.oracle.com/technetwork/articles/servers-storage-admin/sol-why-os-matters-1961737.html.
1) The types of queries that benefit from parallel processing include select statements that scrutinize a large number of pages but return comparatively a small number of rows, the create index statements, select statements that consists of order by and union, and the select statements that make use of merge joints. Moreover, the Decision Support System (DSS) queries that way in a large number of tables and outputs the summary particulars gain the majority from the parallel processing (Sharma, Singh & Singh, 2016).
2) The benefits of parallel database include speed because the server divides a particular database request to pieces and sends out each piece to distinct computer. Then works is done on the pieces concurrently and outcomes are combined which speeds up the procedure. Another benefit of parallel processing is capacity because when new users ask for admittance to the database, the admin puts in more PC’s to parallel sever which increases the overall capacity (Babu, 2012). Along with this, parallel databases are reliable because the shutdown of one PC in the cluster does affect the working. The database senses that a particular PC is not available and reroutes the work to the other PC’s.
3) Oracle utilizes memory to speed up the process using the dual format architecture. In this architecture, the tables are represented concurrently in a usual row arrangement and new in memory column arrangement. In this way, the analytic queries are sent to the row format and the OLTP (Online Transaction Processing Queries) are sent to the column format which utilizes the memory and speeds up the process with excellent performance.
Babu, S. (2012). Massively Parallel Databases and MapReduce Systems. Foundations And Trends® In Databases, 5(1), 1-104. doi: 10.1561/1900000036
Sharma, M., Singh, G., & Singh, R. (2016). Design and analysis of stochastic DSS query optimizers in a distributed database system. Egyptian Informatics Journal, 17(2), 161-173. doi: 10.1016/j.eij.2015.10.003