M2   Designing Resilient Distributed Systems - High Availability 
Evan Marcus, Veritas Software; Hal Stern, Sun Microsystems, Inc.

Who should attend: Novice and advanced UNIX system and network administrators, and UNIX developers concerned with building applications that can be deployed and managed in a highly resilient manner. A basic understanding of UNIX system programming, UNIX shell programming, and network environments is required.

This course will explore procedures and techniques for designing, building and managing predictible, resilient UNIX-based systems in a distributed environment. Hardware redundancy, system redundancy, monitoring and verification techniques, network implications, system and application programming issues will all be addressed. We will discuss the trade-offs between cost, reliability and complexity. Topics to be covered:

	What is high availability? Who does and does not need it?
	Defining uptime and cost; "big rules" of system design
	Disk and data redundancy; RAID and SCSI arrays
	Host redundancy in HA configurations
	Network dependencies
	Application system programming concerns
	Anatomy of failovers: applications, systems, management tools
	Planning disaster recovery sites and data updates
	Security implications
	Upgrade and patch strategies
	Backup systems: off-site storage, redundancy and disaster recovery issues
	Managing the system managers, processes, verification

 

Hal Stern  (S2, M2)  is a Distinguished Systems Engineer with Sun Microsystems, Inc., where he focuses on high-end server technology, operations management, networking, performance tuning, and information systems architecture. Hal has been a UNIX administrator and user for more than 10 years. Before joining Sun, he developed molecular modeling software for a Boston area start-up company and was on the research staff at Princeton University. He is the author of Managing NFS & NIS and the author of several articles on application performance and network design.