dipartimento di elettronica e informazione novel, emerging computing system technologies smart...
TRANSCRIPT
DIPARTIMENTO DI ELETTRONICA E INFORMAZIONE
Novel,EmergingComputing System Technologies
Smart Technologies Smart Technologies for Effective Reconfiguration:for Effective Reconfiguration:The FASTER approachThe FASTER approach
May 29th – 31st 2013International Conference on IC Design and Technology
Pavia, Italy
M. D. Santambrogio, C. Pilato, D. Pnevmatikatos, K. Papadimitriou, D. Stroobandt, D. Sciuto
http://www.fp7-faster.eu/
Reconfigurable TechnologyReconfigurable Technology
• Technology for adaptable hardware systems Can add/remove components at
run-time/product lifetime Flexibility at hardware speed (not quite ASIC) Parallelism at hardware level (depending on
application) Ideally: alter function & interconnection of
blocks• Implementation in:
FPGAs: fine grain, complex gate plus memory and DSP blocks
Coarse Grain (custom) chips: multiple ALUs, multiple (simple) programmable processing blocks, etc.
2
An issue as a new opportunityAn issue as a new opportunity
• Programming has become very difficult Impossible to balance all constraints manually
3
• More computational horse-power than ever before Cores are free, reconfigurable logic available on chip,
cores can be heterogeneous• Energy is new constraint
Software must become energy and space aware
• Modern computing systems need to be flexible and adaptive To optimize and meet their requirements taking
advantage as much as possible of the underlying complex heterogeneous architectures
FASTER MotivationFASTER Motivation
• Creating reconfigurable systems is not straightforward! Reconfiguration cost is substantial (use wisely) Tool support for these tasks is still quite basic Resource management is up to the user
The designer has to: Identify portions to be reconfigured Establish a schedule that (a) respects
dependencies (b) achieves performance and other constraints
Manage the system resources (also at run-time) Verify a changing system!
4
FASTER Goals and InnovationFASTER Goals and Innovation
• Include reconfigurability as an explicit design concept in computing systems design, along with methods and tools that support run-time reconfiguration in the entire design methodology Provide a framework for analysis, synthesis
and verification of a reconfigurable system Provide efficient and transparent runtime
support for partial and dynamic reconfiguration, including micro-reconfiguration
• Demonstrate usability & performance on commercial applications and platforms (Maxeler, ST Microelectronics, Synelixis)
5
FASTER PlatformsFASTER Platforms
• Bridging the gap between HPC and embedded systems Opportunities and challenges of reconfiguration
in both the domains
• High-Performance Computing Systems Maxeler MPC MaxWorkstation
• FPGA-based Embedded Systems Xilinx University Program Board (XUPV5-
LX110T) AVNET Zedboard (SoC XC7Z020)
6
FASTER: Overall MethodologyFASTER: Overall Methodology
7
Design Phase and Runtime Design Phase and Runtime SupportSupport
• Define a reconfiguration-aware design methodology that exploits FPGAs: Generate hardware and software components (including runtime support) on the
top of existing vendor flows• Exploit dynamic reconfigurability for different target reconfigurable
architectures. Both HPC and embedded systems
• Define and implement a new generation of self reconfigurable architectures based on Linux
8
System analysis and designSystem analysis and design
9
.c
Task Graph Generation
High Level Analysis
DFG Extraction
Static Baseline Scheduling
Partitioningand Optimizations
Run-time Support and Verification
.xml
.c .xml
- Architecture- Additional application
information
- Annotated source code(C+OpenMP)
- Source code for CPU- DFGs for HW blocks- Mapping Configurations
Mapping and Floorplanning
- HLS- System generation
.xml
Identifying Level of Identifying Level of ReconfigurabilityReconfigurability
• Assigning each task of the application to the “best” processing element Reconfiguration is implicitly considered
• Based on a metaheuristic iterative algorithm
10
Objectives: function of occupation area, execution time, power, number of reconfigurations etc...
T1
T2 T3 T4
T5
Architectural TemplateXML
MAP MappingMAP
Iterative, multi objectives:-Runtime-Power-Area-…
Convergence
Library XML
Micro-reconfiguration Micro-reconfiguration OptimizationOptimization
• In some applications we can identify hardware accelerators with slow‐changing “parameters” Filter coefficients
• Parameters trigger a small-scale reconfiguration
• Design of cores based on Tunable FPGA blocks: Identify parameters Create bitfile with “holes” Parameter values => reconfiguration bits for
missing “holes” Fine grain, faster reconfiguration time!
11
Verifying Reconfigurable Verifying Reconfigurable SystemsSystems
• Study design validation approaches: simulation, emulation and formal verification
• Extend symbolic simulation to dynamic aspects of reconfigurable design
• In some cases static approaches may not be able to verify the entire RC system We use run‐time verification. Address and
minimize impact on:• Speed, area and power• Light‐weight architectural support
12
Run-time SystemRun-time System
• Evaluate reconfiguration overhead
• Propose advanced mechanisms to support Scheduling Dynamic reconfiguration (including micro-
reconfiguration) Run-time verification
• Provide run-time support for dynamic reconfiguration based on static analysis Extension of OS capabilities Efficient on-line scheduling and placement of
task modules
13
OS-based ManagementOS-based Management
• Provide software support for dynamic partial reconfiguration on a Linux-based operating systems Reconfiguration process managed from the
OS in a transparent way
• Hardware-independent interface for software developers based on the GNU/Linux Addition and removal of reconfigurable
components
• Easier programming interface for specific drivers OS customization for specific architectures
14
Expected Results and ConclusionsExpected Results and Conclusions
• FASTER is a focused project that builds on combined partners expertise as well as on past research work and projects
• We focus on (and hope to demonstrate): productivity improvement in
implementation and verification of dynamically changing systems
total ownership cost reduction (NIDS and RTM systems)
performance improvement under power constraints for Global Illumination and Image Analysis application
15
Challenges & Challenges & OpportunitiesOpportunities
• Tool support for analysis and system definition
• Specification of changing system(s)• Reconfigurable granularity:
influenced by tools and applications• Architectural support for
reconfiguration (vendor?)• Metrics: include design effort/time,
total ownership cost
16