A statistical model of skewed-associativity

This paper presents a statistical model for explaining why skewed-associativity removes conflicts better than setassociativity. We show that, with a high probability, 2-way skewed associativity emulates full associativity for workingsets up to half the cache size, and we show that 3-way skewed-associativity is almost equivalent to full associativity.

• Type: Conference
• Year: 2003
• Where: ISPASS
• Authors: Pierre Michaud
• Keywords: 2-way Skewed Associativity, ISPASS 2003, Skewed-associativity Removes Conflicts, Software Engineering, Statistical Model