User Tools

Site Tools


buzzword

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

buzzword [2019/12/05 11:53] alsermbuzzword [2019/12/14 20:25] (current) – external edit 127.0.0.1
Line 253: Line 253:
     * BitWeaving     * BitWeaving
   * Computing Architectures with Minimal Data Movement   * Computing Architectures with Minimal Data Movement
-  * Mindset on reviewing manuscripts and scientific process  
-  * Suggestions on critical paper review 
-  * Mindset issues everywhere 
-    * Bandwidth bottleneck in Zurich Airport 
-    * Wrong methodology in design space exploration: Building bridges across Manhattan and Brooklyn 
   * 3D-Stacked Logic+Memory   * 3D-Stacked Logic+Memory
   * Logic Layer    * Logic Layer 
Line 683: Line 678:
   * Collaborative patterns   * Collaborative patterns
  
 +===== Lecture 18 (22.11 Fri.) =====
 +  * Instruction prefetching
 +  * Data prefetching
 +  * Memory Hierarchy
 +  * Memory Read/Write Latency
 +  * Memory Bandwidth
 +  * Memory Footprint
 +  * Caches as Bandwidth Filters
 +  * Little's Law
 +  * Occupancy
 +  * Latency
 +  * Throughput
 +  * Queueing Resources
 +  * Compulsory Miss
 +  * Demand Miss
 +  * Spatial and Temporal Locality
 +  * Fetch Granule
 +  * Hardware Prefetching
 +  * Software Prefetch Instruction
 +  * Code Reordering
 +  * Speculative Execution
 +  * Loop Unrolling
 +  * Load Hoisting
 +  * Prefetch Degree
 +  * Three Prefetch Metrics
 +    * Accuracy
 +    * Coverage
 +    * Timeliness
 +  * Heuristic-Based Next-N-Line Prefetching
 +  * History-Based Target Line Prefetching
 +  * Heuristic-Based Wrong-Path Prefetching
 +  * Hybrid Prefetching
 +  * Branch Predictor
 +  * Branch Target Buffer (BTB)
 +  * Next-Line Prefetchers
 +  * Stride Prefetchers
 +  * Cache-Block Address Based Stride Prefetching
 +  * Correlation-Based Prefetchers
 +  * Content-Birected Prefetchers
 +  * Precomputation or Execution-Based prefetchers
 +  * Address Correlation Based Prefetching
 +  * Markov Model and Markov Prefetchers
 +  * Prefetch Confidence
 +  * Hybrid Hardware Prefetchers
 +  * Execution-based Prefetchers
 +  * Speculative Thread
 +  * Feedback-Directed Prefetcher
 +  * Prefetcher Throttling
 ===== Lecture 19 (28.11 Thu.) ===== ===== Lecture 19 (28.11 Thu.) =====
  
Line 733: Line 776:
   * Crash consistency   * Crash consistency
   * Checkpointing   * Checkpointing
-  * Flynn´s taxonomy of computers+  * Flynn'taxonomy of computers
   * Parallelism   * Parallelism
   * Performance   * Performance
Line 742: Line 785:
   * Data parallelism   * Data parallelism
   * Task level parallelism   * Task level parallelism
 +  * Utilization
 +  * Redundancy
 +  * Efficiency
 +  * Amdahl's law
 +  * Bottlenecks in parallel portion
   * Multiprocessor   * Multiprocessor
-  * Loosely coupled +  * Loosely coupled multiprocessors 
-  * Tightly coupled+  * Tightly coupled multiprocessors
   * Shared global memory address space   * Shared global memory address space
   * Shared memory synchronization   * Shared memory synchronization
-  * Cache coherence 
-   
- 
-===== Lecture 21b (5.12 Thu.) ===== 
- 
-  * Memory consistency / memory ordering 
-  * Shared resource management 
   * Interconnects   * Interconnects
   * Programming issues in tightly coupled multiprocessor   * Programming issues in tightly coupled multiprocessor
Line 759: Line 800:
   * Linear speedup   * Linear speedup
   * Superlinear speedup   * Superlinear speedup
 +  * Shared resource management
   * Unfair comparison   * Unfair comparison
   * Cache/memory effect   * Cache/memory effect
-  * Utilization + 
-  * Redundancy +   
-  * Efficiency + 
-  * Amdahl's law +===== Lecture 21b (5.12 Thu.) ===== 
-  * Bottlenecks in parallel portion+ 
 +  * Memory consistency / memory ordering
   * Ordering of operations   * Ordering of operations
 +  * Local ordering
 +  * Global ordering 
   * Sequential consistency   * Sequential consistency
   * Weaker memory consistency   * Weaker memory consistency
   * Memory fence instructions   * Memory fence instructions
 +  * Consequences of Sequential Consistency
 +  * Issues with Sequential Consistency
 +  * Global order requirement 
 +  * Aggressiveness
 +  * Out-of-order execution 
   * Higher performance   * Higher performance
   * Burden on the programmer   * Burden on the programmer
-  * Coherence scheme+  * Mutual exclusion 
 +  * Protecting shared data 
 +  * Ease of debugging 
 +  * Correctness 
 +  * MIMD processor 
 +  * Dataflow processor 
 +   
 + 
 +===== Lecture 22 (6.12 Fri.) ===== 
 + 
 +  * Cache coherence 
 +  * Memory consistency 
 +  * Shared memory model 
 +  * Software coherence 
 +    * Coarse-grained (page-level) 
 +    * Non-cacheable 
 +    * Fine-grained (cache flush) 
 +  * Hardware coherence
   * Valid/invalid   * Valid/invalid
   * Write propagation   * Write propagation
   * Write serialization   * Write serialization
   * Update vs. Invalid   * Update vs. Invalid
-  * Cache coherence 
   * Snoopy bus   * Snoopy bus
   * Directory   * Directory
-  Directory optimizations +    Exclusive bit 
-  * Directory bypassing+  * Directory optimizations (bypassing)
   * Snoopy cache   * Snoopy cache
   * Shared bus   * Shared bus
Line 791: Line 857:
   * Broadcast   * Broadcast
   * Bus request   * Bus request
-  * Downgrade +  * Downgrade/upgrade
-  * Upgrade+
   * Snoopy invalidation   * Snoopy invalidation
   * Cache-to-cache transfer   * Cache-to-cache transfer
Line 809: Line 874:
   * Shared-data-structure   * Shared-data-structure
   * Token Coherence   * Token Coherence
-  * Virtual bus+  * Coherence for NDAs 
 +  * Optimistic execution 
 +  * Signature 
 +  * Commit/re-execute 
 + 
 +===== Lecture 23 (12.12 Thu.) ===== 
 +  * Interconnects  
 +  * Cache coherence 
 +  * Interconnect networks:  
 +      * Topology 
 +      * Routing 
 +      * Buffering and flow control 
 +          * Oversubscription of routers 
 +  *  Terminology: 
 +      * Network Interface 
 +      * Link 
 +      * Switch/router 
 +      * Channel 
 +      * Node 
 +      * Message 
 +      * Packet 
 +      * Flit 
 +      * Direct/Indirect network 
 +  * Properties of a Network Topology: 
 +      * Regular/Irregular 
 +      * Routing distance 
 +      * Diameter 
 +      * Average distance 
 +      * Bisection Bandwidth 
 +      * Blocking/non-blocking. Rearrangeable non-blocking 
 +  * Topologies: 
 +      * Bus 
 +      * P2P 
 +      * Crossbar 
 +      * Ring 
 +      * Tree 
 +      * Omega 
 +      * Hypercube 
 +      * Mesh 
 +      * Torus 
 +      * Butterfly 
 +  * cost, latency, contention, energy, bandwidth, overall performance 
 +  * Circuit switching network 
 +  * Multistage network  
 +  * Fetch-and-add 
 +  * Unidirectional Ring 
 +  * Bidirectional rings 
 +  * Hierarchical rings 
 +  * Mesh: asymmetricity on the edge 
 +  * Torus 
 +  * H-tree 
 +  * Fat-tree 
 +  * Hyper-cube. Cosmic Cube 
 +  * Routing mechanism: Arithmetic, Source-based, Table-based lookup 
 +  * Types of routing algorithm: deterministic, oblivious, adaptive 
 +  * Deadlock 
 +  * Oblivious routing 
 +  * Adaptive Routing: minimal adaptive, non-minimal adaptive 
 +  * Flow Control 
 +  * Handling contention: buffer, drop or misroute 
 +  * Flow control methods: 
 +      * Circuit switching 
 +      * Bufferless 
 +      * Store and forward 
 +      * Virtual cut through 
 +      * Wormhole 
 +  * Performance and congestion at high loads 
 +  * Store-and-forward 
 +  * Cut-though flow control 
 +  * Wormhole flow control 
 +  * Head of Line Blocking
  
-===== Lecture 21c (5.12 Thu.) =====+===== Lecture 24 (13.12 Fri.) ===== 
 +  * Load latency curve 
 +  * Performance of interconnection networks 
 +  * On-chip networks 
 +  * Difference between off-chip and on-chip networks 
 +  * Network buffers 
 +  * Efficient routing 
 +  * Advantages of on-chip interconnects 
 +  * Pin constraints 
 +  * Wiring resources 
 +  * Disadvantages of on-chip interconnects 
 +  * Energy/power constraint 
 +  * Tradeoffs of interconnect design 
 +  * Buffers in NoC routers 
 +  * Bufferless routing 
 +  * Flit-level routing 
 +  * Deflection routing 
 +  * Buffer and link energy consumption 
 +  * Self-throttling 
 +  * Livelock freedom problem 
 +  * Golden packet for livelock freedom 
 +  * Reassembly buffers 
 +  * Packet retransmission 
 +  * Packet scheduling
buzzword.1575546830.txt.gz · Last modified: 2019/12/05 11:53 by alserm