{"id":{"repo_id":"mcmaster","oai_identifier":"oai:macsphere.mcmaster.ca:11375/6357"},"canonical_url":"https://search.ndltd.org/etd/mcmaster/oai:macsphere.mcmaster.ca:11375/6357","repository":{"repo_id":"mcmaster","name":"McMaster University","base_url":"https://macsphere.mcmaster.ca/server/oai/request"},"display":{"title":"Performance Optimization of Hierarchical Memory Systems","abstract":"<p>The gap between processor speeds and memory speeds is increasing. The performance of supercomputers and the scalability of multiprocessor systems is very dependent on the memory system speed.</p> <p>A cache system helps to narrow the processor/memory speed gap, but cannot completely decouple the processor from slow memory.</p> <p>The optimization of main memory performance and the use of a deep multilevel cache hierarchy are proposed here to bridge the processor/memory latency gap.</p> <p>A novel design that combines optimized bank interleaving with several main memory (DRAM) timing modes to increase memory performance is presented. Four different protocols based on this design are proposed and investigated.</p> <p>Enforcing the inclusion property for multi-level caches is proposed. A new design that uses three level caches is preserved and three different models are given.</p> <p>A design flow graph that makes the design of a multi-level memory system simpler and more flexible is introduced. Selected traces that match real workloads running on a wide range of computers are used to calculate realistic overall system performance.</p>","abstract_html":"&lt;p&gt;The gap between processor speeds and memory speeds is increasing. The performance of supercomputers and the scalability of multiprocessor systems is very dependent on the memory system speed.&lt;/p&gt; &lt;p&gt;A cache system helps to narrow the processor/memory speed gap, but cannot completely decouple the processor from slow memory.&lt;/p&gt; &lt;p&gt;The optimization of main memory performance and the use of a deep multilevel cache hierarchy are proposed here to bridge the processor/memory latency gap.&lt;/p&gt; &lt;p&gt;A novel design that combines optimized bank interleaving with several main memory (DRAM) timing modes to increase memory performance is presented. Four different protocols based on this design are proposed and investigated.&lt;/p&gt; &lt;p&gt;Enforcing the inclusion property for multi-level caches is proposed. A new design that uses three level caches is preserved and three different models are given.&lt;/p&gt; &lt;p&gt;A design flow graph that makes the design of a multi-level memory system simpler and more flexible is introduced. Selected traces that match real workloads running on a wide range of computers are used to calculate realistic overall system performance.&lt;/p&gt;","abstract_has_math":false,"creators":["Mekhiel, Nassief Nagi"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":"Electrical and Computer Engineering","school":null,"contributors":[],"advisors":["McCrackin, Daniel"],"committee_chairs":[],"committee_members":[],"year":1995,"date_issued":"1995-09","date_published":"1995-09","updated_at":"2026-11-18T19:37:34Z","subjects":["Electrical and Computer Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://doi.org/10.71548/18943"],"render_values":[{"text":"https://doi.org/10.71548/18943","href":"https://doi.org/10.71548/18943","code":true}]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["opendissertations/1672","3229","1368121"],"render_values":[{"text":"opendissertations/1672","href":null,"code":true},{"text":"3229","href":null,"code":true},{"text":"1368121","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/11375/6357","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"http://hdl.handle.net/11375/6357","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://macsphere.mcmaster.ca/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Amacsphere.mcmaster.ca%3A11375%2F6357","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["McCrackin, Daniel"]},{"key":"dc:contributor.department","label":"Department","values":["Electrical and Computer Engineering"]},{"key":"dc:creator","label":"Author","values":["Mekhiel, Nassief Nagi"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2014-06-18T16:35:11Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2014-06-18T16:35:11Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2010-06-22"]},{"key":"dc:date.issued","label":"Date","values":["1995-09"]},{"key":"dc:type","label":"Dc Type","values":["thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Electrical and Computer Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["opendissertations/1672","3229","1368121"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/11375/6357","https://doi.org/10.71548/18943"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["<p>The gap between processor speeds and memory speeds is increasing. The performance of supercomputers and the scalability of multiprocessor systems is very dependent on the memory system speed.</p> <p>A cache system helps to narrow the processor/memory speed gap, but cannot completely decouple the processor from slow memory.</p> <p>The optimization of main memory performance and the use of a deep multilevel cache hierarchy are proposed here to bridge the processor/memory latency gap.</p> <p>A novel design that combines optimized bank interleaving with several main memory (DRAM) timing modes to increase memory performance is presented. Four different protocols based on this design are proposed and investigated.</p> <p>Enforcing the inclusion property for multi-level caches is proposed. A new design that uses three level caches is preserved and three different models are given.</p> <p>A design flow graph that makes the design of a multi-level memory system simpler and more flexible is introduced. Selected traces that match real workloads running on a wide range of computers are used to calculate realistic overall system performance.</p>"]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Doctor of Philosophy (PhD)"]},{"key":"dc:title","label":"Title","values":["Performance Optimization of Hierarchical Memory Systems"]},{"key":"others:access_status","label":"Others Access Status","values":["open.access"]}]}],"canonical_facts":{"dc:contributor.advisor":["McCrackin, Daniel"],"dc:contributor.department":["Electrical and Computer Engineering"],"dc:creator":["Mekhiel, Nassief Nagi"],"dc:date.accessioned":["2014-06-18T16:35:11Z"],"dc:date.available":["2014-06-18T16:35:11Z"],"dc:date.created":["2010-06-22"],"dc:date.issued":["1995-09"],"dc:description.abstract":["<p>The gap between processor speeds and memory speeds is increasing. The performance of supercomputers and the scalability of multiprocessor systems is very dependent on the memory system speed.</p> <p>A cache system helps to narrow the processor/memory speed gap, but cannot completely decouple the processor from slow memory.</p> <p>The optimization of main memory performance and the use of a deep multilevel cache hierarchy are proposed here to bridge the processor/memory latency gap.</p> <p>A novel design that combines optimized bank interleaving with several main memory (DRAM) timing modes to increase memory performance is presented. Four different protocols based on this design are proposed and investigated.</p> <p>Enforcing the inclusion property for multi-level caches is proposed. A new design that uses three level caches is preserved and three different models are given.</p> <p>A design flow graph that makes the design of a multi-level memory system simpler and more flexible is introduced. Selected traces that match real workloads running on a wide range of computers are used to calculate realistic overall system performance.</p>"],"dc:description.degree":["Doctor of Philosophy (PhD)"],"dc:identifier.other":["opendissertations/1672","3229","1368121"],"dc:identifier.uri":["http://hdl.handle.net/11375/6357","https://doi.org/10.71548/18943"],"dc:subject":["Electrical and Computer Engineering"],"dc:title":["Performance Optimization of Hierarchical Memory Systems"],"dc:type":["thesis"],"others:access_status":["open.access"]},"updated_at":"2026-11-18T19:37:34Z"}