User Tools

Site Tools


buzzword

Differences

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

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
buzzword [2018/11/28 14:24] – [Lecture 19b (28.11 Thu.)] yaglikcabuzzword [2019/09/20 11:09] (current) juang
Line 63: Line 63:
   * Error Correcting Codes (ECC)   * Error Correcting Codes (ECC)
  
-===== Lecture 2 (20.09 Wed.) =====+===== Lecture 2 (20.09 Thu.) =====
   * Rowhammer   * Rowhammer
   * Memory reliability   * Memory reliability
Line 820: Line 820:
    * Heterogeneity of CPU and GPUs    * Heterogeneity of CPU and GPUs
    * Predictability and robustness    * Predictability and robustness
 +
 +===== Lecture 20 (29.11 Thu.) =====
 +  * DRAM
 +  * NVM
 +  * Flash
 +  * Processing in Memory
 +  * Hardware Security
 +  * Heterogeneous Multi-Core Systems
 +  * Bottleneck Acceleration
 +  * Heterogeneity (Asymmetry)
 +  * Symmetric design 
 +  * One-size-fits-all
 +  * Quality of Service (QoS)
 +  * Hybrid Memory Controllers
 +  * Heterogeneous agents (e.g., CPUs, GPUs, and HWAs)
 +  * Heterogeneous memories: Fast vs. Slow DRAM
 +  * Heterogeneous interconnects: Control, Data, Synchronization
 +  * Amdahl’s Law
 +  * Synchronization overhead
 +  * Load imbalance overhead
 +  * Resource sharing overhead
 +  * Sequential portions (Amdahl’s “serial part”)
 +  * Critical sections
 +  * Barriers
 +  * Asymmetric Chip Multiprocessor (ACMP)
 +  * Bottleneck Acceleration
 +  * Staged Execution
 +  * Data Marshaling
 +  * Phase Change Memory
 +
 +===== Lecture 21 (05.12 Wed.) =====
 +  * GPU
 +  * Programming model
 +  * Sequential
 +  * SIMD
 +  * SPMD
 +  * SIMT
 +  * Warp (wavefront)
 +  * Multithreading of warps
 +  * Warp-level FGMT
 +  * Latency-hiding
 +  * Interleave warp execution
 +  * Registers of thread ID
 +  * Warp-based SIMD vs. Traditional SIMD
 +  * GPGPU programming
 +  * Inherent parallelism
 +  * Data parallelism
 +  * GPU main bottlenecks
 +  * CPU-GPU data transfers
 +  * DRAM memory
 +  * Task offloading
 +  * Serial code (host)
 +  * Parallel code (device)
 +  * Bulk synchronization
 +  * Transparent scalability
 +  * Memory hierarchy
 +  * Indexing and memory access
 +  * Streaming multiprocessor (SM)
 +  * Streaming processor (Vector lane)
 +  * Occupancy
 +  * Memory coalescing
 +  * Shared memory tiling
 +  * Bank conflict
 +  * Padding
 +  * SIMD utilization
 +  * Atomic operations
 +  * Histogram calculation
 +  * CUDA streams
 +  * Asynchronous transfers
 +  * Heterogeneous systems
 +  * Unified memory
 +  * System-wide atomic operations
 +  * Collaborative computing
 +  * CPU+GPU collaboration
 +  * Collaborative patterns
 +      * Data partitioning
 +      * Task partitioning
 +          * Coarse-grained
 +          * Fine-grained
 +  * Bézier surfaces
 +  * NVIDIA Pascal
 +  * NVIDIA Volta
 +  * Padding
 +  * Chai benchmark suite
 +
 +===== Lecture 22 (6.12 Thu.) =====
 +  * Persistent memory
 +  * Crash consistency
 +  * Checkpointing
 +  * Flynn´s taxonomy of computers
 +  * Parallelism
 +  * Performance
 +  * Power consumption
 +  * Cost efficiency
 +  * Dependability
 +  * Instruction level parallelism
 +  * Data parallelism
 +  * Task level parallelism
 +  * Multiprocessor
 +  * Loosely coupled
 +  * Tightly coupled
 +  * Shared global memory address space
 +  * Shared memory synchronization
 +  * Cache coherence
 +  * Memory consistency
 +  * Shared resource management
 +  * Interconnects
 +  * Programming issues in tightly coupled multiprocessor
 +  * Sublinear speedup
 +  * Linear speedup
 +  * Superlinear speedup
 +  * Unfair comparison
 +  * Cache/memory effect
 +  * Utilization
 +  * Redundancy
 +  * Efficiency
 +  * Amdahl's law
 +  * Bottlenecks in parallel portion
 +  * Ordering of operations
 +  * Sequential consistency
 +  * Weaker memory consistency
 +  * Memory fence instructions
 +  * Higher performance
 +  * Burden on the programmer
 +  * Coherence scheme
 +  * Valid/invalid
 +  * Write propagation
 +  * Write serialization
 +  * Update vs. Invalid
 +  * Cache coherence
 +  * Snoopy bus
 +  * Directory
 +  * Directory optimizations
 +  * Directory bypassing
 +  * Snoopy cache
 +  * Shared bus
 +  * VI protocol
 +  * MSI (Modified, Shared, Invalid)
 +  * Exclusive state
 +  * MESI (Modified, Exclusive, Shared, Invalid)
 +  * Illinois Protocol (MESI)
 +  * Broadcast
 +  * Bus request
 +  * Downgrade
 +  * Upgrade
 +  * Snoopy invalidation
 +  * Cache-to-cache transfer
 +  * Writeback
 +  * MOESI (Modified, Owned, Exclusive, Shared, Invalid)
 +  * Directory coherence
 +  * Race conditions
 +  * Totally-ordered interconnect
 +  * Directory-based protocols
 +  * Set inclusion test
 +  * Linked list
 +  * Bloom filters
 +  * Contention resolution
 +  * Ping-ponging
 +  * Synchronization
 +  * Shared-data-structure
 +  * Token Coherence
 +  * Virtual bus
 +
 +===== Lecture 23 (12.12 Wed.) =====
 +  * Interconnection Network, Interconnect
 +  * Topology
 +  * Routing
 +  * Buffering and Flow Control
 +  * Switch/Router
 +  * Channel
 +  * Wire
 +  * Packet
 +  * Path
 +  * Bus
 +  * Mesh, 2D Mesh
 +  * Throttling
 +  * Oversubscription
 +  * Network Interface
 +  * Link
 +  * Node
 +  * Message
 +  * Flit
 +  * Direct/Indirect Network
 +  * Radix
 +  * Regular/Irregular Topology
 +  * Routing Distance
 +  * Diameter
 +  * Bisection Bandwidth
 +  * Congestion
 +  * Blocking/non-blocking Interconnect
 +  * Crossbar
 +  * Ring
 +  * Tree
 +  * Omega
 +  * Hypercube
 +  * Torus
 +  * Butterfly
 +  * Arbitration
 +  * Point-to-Point
 +  * Multistage Network
 +  * Hop
 +  * Circuit Switching
 +  * Packet Switching
 +  * Tree saturation
 +  * Deadlock
 +  * Circular dependency
 +  * Oblivious Routing
 +  * Adaptive Routing
 +  * Packet Format
 +  * Header
 +  * Payload
 +  * Error Code
 +  * Virtual Channel Flow Control
 +
 +===== Lecture 24 (13.12 Thu.) =====
 +  * 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.1543415045.txt.gz · Last modified: 2019/02/12 16:33 (external edit)