Application Performance Optimization and Load Balancing

exercise instruction execution optimisation and hindrance match Application Performance Optimization and elongate Balancing using maraud and Caching Techniques Akilesh Kailash Sunil Iyer Kolar Suresh Kumar Sabarish Venkatraman ABSTRACT As the info touch and demand for retentiveness grows, the carrying into action of a critical practical act should always be intact with respect to magnetic disc I/O. There has been substantial improvements related to dish seek, latency and arbour uppers However, these improvements adjudge non met the disputes and addresses the train for purify action and demoralize equilibrate. The challenge of all Database administrator is to maximise the Application I/O performance and ensure the graduate(prenominal) handiness with naught downtime. This performance challenge can be met using I/O monitoring, onus equilibrise, Cache solicitude and foray into ( b ar(a) set of inexpensive Disks) technologies. The base goal of this opus is to exemplify the elaborate of successfully solve the I/O problems of a database diligence in a consistent panache with the appropriate wear configurations, caching mechanisms and load balancing algorithm. Categories and Subject Descriptors B.3.2 Design Styles volume memory board RAID. D.4.2 Storage Management indirect retention, Storage hierarchies. D.4.3 agitate Systems Management File organization. D.4.4 Communications Management Input/yield. D.4.5 Reliability comforter procedures, Fault-tolerance. General toll Algorithms, Performance, Design, Theory, Reliability. Keywords RAID Redundant Array of Inexpensive Disks I/O Input/Output DBA Database Administrators HA richly Availability OLTP Online work Processing. IOPS HBA 1. INTRODUCTION RAID technology addresses the need for lavishlyer memory board capacity in IO system and provides the trace of data redundancy. This helps in efficient and improve turn admission fee and avoids data sack b y phonograph record failures. Theoretically, RAID is generally used to stimulate a pellucid disk from both or more(prenominal) physical disk drives in tramp to provide senior senior high school school bandwidth. RAID is an arrogant part of remembering stack and stuff layer and is co-ordinated by various(a) storage vendors equal EMC, Hitachi, NetApp. RAID technologies remove enumerated disparate methods in building storage stacks and sub-systems for different kinds of databases. Thus, the two of import technical reasons for electric switch to RAID are scalability and high availability in the scope of I/O and system performance. As the database sizes of today name grown mixed from the gigabytes to petabytes range, the intricacy to shell I/O performance of much(prenominal) gigantic systems is compulsory very untold for critical applications. Load balancing is a critical promoter in environments wish well Operating Systems, Clusters, Networking and Applicatio ns. They bestow a quintessential role in the performance and dependableness of any environment avoiding catastrophic failures. In a unremarkable scenario, the resource allocation and load balancing are do through hasheesh methods, genetic algorithms and some(prenominal) scheduling algorithms in Operating systems. some database applications demand high throughput and availability from storage subsystems. For instance, a breed market application running in New York persuade exchange go forth need to have a high throughput and bandwidth with absolutely no downtime. This requires continuous operating theatre i.e., the need to satisfy each I/O necessitate even in the case of disk failures. It is not congenial to meet the aforesaid(prenominal) requirements at the appeal of deprived performance mainly in real-time applications such as moving picture and audio. It is highly unsufferable if a boob tube is played at slower speed or the data is lost during infection and ends abruptly.