COMP25212
System Architecture
10
Semester 2
Undergraduate
Level 2
Davide Bertozzi
Unit coordinator
Department of Computer Science
1
FHEQ level (Framework for Higher Education Qualifications) ' Middle part of Bachelors '
European Credit Transfer & Accumulation System Rating : 5.0
<p>This course introduces the key ideas behind the design of modern computer processors and explains how advanced architectural techniques are used to improve performance and efficiency. Students will explore why these techniques are needed and how they are applied in real computing systems.</p><p><strong>This course covers the following topics and themes:</strong></p><p><strong>• Why modern processors need advanced design techniques </strong><br>Performance limitations in modern computing systems and the motivation for advanced processor architecture techniques.</p><p><strong>• Reducing delays when accessing memory </strong><br>Memory latency and the role of caching, including cache organisation and operation.</p><p><strong>• Overlapping work inside the processor </strong><br>Instruction pipelining, processor throughput, superscalar architectures.</p><p><strong>• Handling dependencies and decision points during execution </strong><br>Pipeline limitations and hazards, including data and control hazards, and techniques to improve pipeline performance.</p><p><strong>• Running multiple tasks at once on a single processor </strong><br>Multi-threading concepts, processor support for multi-threading, and simultaneous multi-threading.</p><p><strong>• Using multiple processor cores </strong><br>Motivation for multi-core processors and possible multi-core processor organisations.</p><p><strong>• Sharing data safely between processing units </strong><br>Data sharing, cache coherence, and consistency in multi-core systems. </p>
<p>This course introduces the key ideas behind the design of modern computer processors and explains how advanced architectural techniques are used to improve performance and efficiency. Students will explore why these techniques are needed and how they are applied in real computing systems.</p><p><strong>This course covers the following topics and themes:</strong></p><p><strong>• Why modern processors need advanced design techniques </strong><br>Performance limitations in modern computing systems and the motivation for advanced processor architecture techniques.</p><p><strong>• Reducing delays when accessing memory </strong><br>Memory latency and the role of caching, including cache organisation and operation.</p><p><strong>• Overlapping work inside the processor </strong><br>Instruction pipelining, processor throughput, superscalar architectures.</p><p><strong>• Handling dependencies and decision points during execution </strong><br>Pipeline limitations and hazards, including data and control hazards, and techniques to improve pipeline performance.</p><p><strong>• Running multiple tasks at once on a single processor </strong><br>Multi-threading concepts, processor support for multi-threading, and simultaneous multi-threading.</p><p><strong>• Using multiple processor cores </strong><br>Motivation for multi-core processors and possible multi-core processor organisations.</p><p><strong>• Sharing data safely between processing units </strong><br>Data sharing, cache coherence, and consistency in multi-core systems. </p>
<p><span style="background-color:rgb(255,255,255);"><span class="TextRun SCXW120207492 BCX0 NormalTextRun" style="-webkit-tap-highlight-color:transparent;-webkit-text-stroke-width:0px;-webkit-user-drag:none;font-family:Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, sans-serif;font-size:11pt;font-style:normal;font-variant-caps:normal;font-variant-ligatures:none !important;font-weight:400;letter-spacing:normal;line-height:18.3458px;margin:0px;orphans:2;padding:0px;text-align:left;text-decoration-color:initial;text-decoration-style:initial;text-decoration-thickness:initial;text-indent:0px;text-transform:none;user-select:text;white-space:normal;widows:2;word-spacing:0px;" data-contrast="auto" xml:lang="EN-GB" lang="EN-GB">This course explains how modern computer processors really work and why they are designed the way they are. Students will explore how techniques such as caching, pipelining, and parallel processing help computers run faster and more efficiently. By understanding these ideas, students will learn how software and hardware interact and how to use computing systems more effectively in practice. </span></span></p>
<p><strong>ILO 1:</strong> Describe the main architectural techniques used in modern processors to improve performance, reliability, and flexibility.</p><p><strong>ILO 2:</strong> Compare the strengths and weaknesses of different processor designs and multi-core organisations.</p><p><strong>ILO 3:</strong> Explain and analyse how caching, pipelining, superscalar, out-of-order and multi-threading techniques affect system performance.</p><p><strong>ILO 4:</strong> Explain the Von Neumann bottleneck and describe design methods to mitigate its impact.</p><p><strong>ILO 5: </strong>Identify and justify suitable processor and system architectures for given performance, reliability, and flexibility requirements.</p><p><strong>ILO 6: </strong>Apply knowledge of architectural techniques to reason about trade-offs between performance, reliability, and resource usage in modern computing systems.</p>
<p>This course introduces the key ideas behind the design of modern computer processors and explains how advanced architectural techniques are used to improve performance and efficiency. Students will explore why these techniques are needed and how they are applied in real computing systems.</p><p><strong>This course covers the following topics and themes:</strong></p><p><strong>• Why modern processors need advanced design techniques </strong><br>Performance limitations in modern computing systems and the motivation for advanced processor architecture techniques.</p><p><strong>• Reducing delays when accessing memory </strong><br>Memory latency and the role of caching, including cache organisation and operation.</p><p><strong>• Overlapping work inside the processor </strong><br>Instruction pipelining, processor throughput, superscalar architectures.</p><p><strong>• Handling dependencies and decision points during execution </strong><br>Pipeline limitations and hazards, including data and control hazards, and techniques to improve pipeline performance.</p><p><strong>• Running multiple tasks at once on a single processor </strong><br>Multi-threading concepts, processor support for multi-threading, and simultaneous multi-threading.</p><p><strong>• Using multiple processor cores </strong><br>Motivation for multi-core processors and possible multi-core processor organisations.</p><p><strong>• Sharing data safely between processing units </strong><br>Data sharing, cache coherence, and consistency in multi-core systems. </p>
<p>Lectures, coursework, online quizzes.</p><p>Guest lecture from industry (TBC).</p><p>Continuous feedback with students will be maintained through active participation in the classroom.</p>
1
Written exam
70%
8
Practical skills assessment
30%
<p><span style="color:black;">Feedback given after assessment</span></p>
COMP15111
Fundamentals of Computer Architecture
Pre-Requisite
Compulsory
COMP15111 is a pre-requisite.<p>COMP15111 is a pre-requisite.</p>
Y
<ol><li>Patterson, David A. (2020). <i>Computer organization and design: the hardware/software interface</i>. Morgan Kaufmann. ISBN:<a href="https://www.librarysearch.manchester.ac.uk/discovery/search?query=isbn,contains,9780128226742&search_scope=MyInst_and_CI&sortby=rank&vid=44MAN_INST:MU_NUI&lang=en&mode=advanced&offset=0" target="_blank"> 9780128226742</a><br> </li><li>Hennessy, John L. (2025). <i>Computer architecture: a quantitative approach</i>. Morgan Kaufmann. ISBN:<a href="https://www.librarysearch.manchester.ac.uk/discovery/search?query=isbn,contains,9780443154065&search_scope=MyInst_and_CI&sortby=rank&vid=44MAN_INST:MU_NUI&lang=en&mode=advanced&offset=0" target="_blank"> 9780443154065 </a></li></ol>
Assessment written exam
2
Lectures
22
Practical classes & workshops
12
0
64