Please fill up the below form
and our Career Consultant will
contact to you in next 12 hours!

Name:


E-mail:


Phone:


Course:


Question / Query / Confusions

Contact icon

Office: 022-24933060
Cell: 9920 436030

Email icon

info@linobox.com
Phone icon Live Practice Labs
test

Our live access labs provide full console access to a self-paced study lab environment. This is an ideal place to re-work lab exercises from class or to experiment with advanced configuration from your home or hotel room. Find out which works for you and get a downloadable eKit when you enroll.



Building High Performance Linux Clusters

Early computers worked with 8 bits, or a byte to solve problems. Most modern computers work with 32 bits at a time, with many dealing with 64 bits per operation, which is similar to increasing the width of a highway.
Another method to increasing computing power will be to increase the clock speed, which is similar to raising the speed limits.
However, there are limits to performance benefits that can be achieved by just increasing the bus width or clock speed. For time-intensive computations, using many computers as oppose to one to solve the same problem, will be an alternate approach.


Who should attend
System administrators, network administrators and software developers who want to implement Cluster.

Prerequisites
Attendees should already have a basic understanding of the Internet and Linux operating systems. You should already be comfortable with editing configuration files and installing software on Linux system. You should understand fundamentals such as hostnames, user-ids, file systems, and file permissions.



Course Outline

* So, what then is a cluster, the different types, their uses and the possible reasons for their wide adoption in industries?
Why consider using Linux as the platform for clustering?

* Talking about HPC architectures; what distinctions can be made between Symmetric Multiprocessors (SMP), Vector processors and Clusters?

* What are those elements you need to consider when designing a solution based on a high performance cluster?

* HPC is mostly centered on the concept of parallelism. How is this important and how does it relate to your application design?

* A file system is a major part of any operating system. In a cluster environment it makes sense to talk about distributed file system. Is NFS or PVFS known to you? What are they and how do they differ from regular file systems e.g. ext3, jfs etc.

* When it comes to managing a cluster in a production environment with a large user base, job scheduling and monitoring are crucial. Can you possibly address this will minimal effort?

* Clusters are built to perform, and you need to know how fast they are. What tool can you use to benchmark a cluster?

PRACTICE LABS:
Building a minimal Linux installation
Setting up a functional cluster using OSCAR
Working with NFS or PVFS as the cluster file system
Handling job-scheduling with OpenPBS
Working with Ganglia for real-time monitoring of CPU and memory usage.
Benchmarking with LINPACK



Trainings
Onsite Training
These courses are available for on-site delivery for groups of 5 or more students. LinoBox offers discounted training for on-site courses of more than 5 students - contact Sanjeet Vanamala at sanjeet@linobox.com or 91-9819586142 to schedule training and arrange details or for any kind of information..