Skip to content
Mar 2008: Building a reliable Internet-scale globally distributed system PDF Print E-mail
Title: Building a reliable Internet-scale globally distributed system

Harald Prokop

Speaker: Harald Prokop, Senior VP,Engineering, Akamai

Date & time: Tuesday, 18th March 2008, 6.30pm - 7.30pm.

Venue: Akamai Technologies - India #77, Jyothi Nivas College Road Koramangala Industrial Layout, Koramangala Bangalore - 560 095 Phone : 6625 1600

Talk Abstract :

Since 1999, Akamai has built a large-scale distributed system with 30,000+ servers spread across 1000+ networks and 70+ countries. Akamai's technology – at its core, applied mathematics and algorithms – has transformed the chaos of the Internet into a predictable, scalable, and secure platform for business and entertainment. Akamai's servers continually monitor the Internet – traffic, trouble spots and overall conditions information – to intelligently optimize routes and deliver content and applications more quickly, reliably, and securely over the Internet. Operating a distributed platform with servers spread across thousands of locations, poses several technical challenges, including maintaining the state and health of the system in real-time, directing user requests to the optimal servers, handling failures, and monitoring and controlling the servers.

This talk will focus on the design principles and methodology that went into architecting the Akamai network and building an Internet-scale reliable system.

About the Speaker:

Harald Prokop is Senior Vice President of Engineering at Akamai. Mr. Prokop manages the Company's engineering efforts and is responsible for building the services that run on Akamai's globally distributed computing platform. He also helps to define and is responsible for executing Akamai's technology strategy with a focus on innovative, market-leading services based on customer requirements.

Mr. Prokop joined Akamai in 1999 as one of Akamai's first engineers and has held various leadership roles within Akamai's Engineering department. Mr. Prokop is one of the architects of Akamai's proprietary system for Internet performance measurement and traffic management, a key component of the Company's services. He designed the development and testing process that brings Akamai services to market.

Prior to Akamai, Mr. Prokop was working on high-performance and parallel computing at the MIT Laboratory for Computer Science. His Master's thesis, "Cache-Oblivious Algorithms," has resulted in an active new research branch of algorithm theory that includes the efficient use of memory hierarchies. Mr. Prokop has also received numerous awards during his career. In 2004, he received Akamai's highest employee recognition, the Daniel Lewin Award, for his outstanding technical contributions to the company. In 1996, Mr. Prokop received a fellowship award from Cusanuswerk, Germany. In 1994, Mr. Prokop was one of the winners of the Bundeswettbewerb Informatik, an annual computer science contest sponsored by the federal government in Germany that attracted thousands of student entrants.

Mr. Prokop received a Master of Science in Electrical Engineering and Computer Science from MIT, and an undergraduate degree in Computer Science from the Universität Karlstruhe in Germany. He has been published in the fields of Internet computing and computer science.

Download Presentation Slides

 
< Prev   Next >
Advertisement

Supporters

IBM-logo.gif