SlideShare una empresa de Scribd logo
1 de 52
Time-driven
reconfiguration-aware
floorplacer
BY
Alessio Montone
alessio.montone@dresd.org
Thesis committee:
S. Dutt (chair), A. Khokhar, M.D. Santambrogio
UIC Thesis Defense: 05/08/2008
2
Rationale and Innovation
Problem statement
Given a reconfigurable architecture, find an on-chip
position for each functional unit
Innovative contribution: taking into account
Target Device Heterogeneity
Target Device reconfiguation capabilities
Inter-FU Communication
3
Aims
Considering the area assignment problem tailored for
reconfigurable architectures, provide
a formalization of the problem, and
an approach (in 3 algorithms) for solving
4
Outline
Introduction
Floorplacement
The Proposed Approach
Experimental Results
Comparison with the state of the art
Conclusions and Future Works
Questions
INTRODUCTION
5
Reconfigurable Architectures - I
On FPGAs
Reconfigurable Devices
Heterogeneous
Reconfiguration Limits
Different types of
Reconfigurable Architectures:
Total
Partial (Static)
Partial (Dynamic)
6
Reconfigurable Architectures - I
Total
7
Reconfigurable Architectures - I
Total
Partial (Static)
8
Reconfigurable Architectures - II
Partial Dynamic
9
Area Assignment Problem
Let consider a Reconfigurable Architecture
Given a scheduled task graph (TG) of the application
Node: Reconfigurable Functional Unit (RFU) [*],
A netlist obtained after post synthesis and technology
mapping (i.e., before placement and routing)
Aim: find an area assignment for each RFU
10
[*] K. Bazargan, R. Kastner, M.S.: 3-d floorplanning: Simulated annealing and greedy placement methods for
reconfigurable computing systems. IEEE Rapid Systems Prototyping (1999)
Related Works - I
[*] introduced the concept of 3D floorplanning for
reconfigurable systems
SA in order to solve HW/SW codesign problem
For each task choose between
HW implementation
SW implementation
Limits
No device limits
considered
No communication
infrastructure
11
[*] K. Bazargan, R. Kastner, M.S.: 3-d floorplanning: Simulated annealing and greedy placement methods for
reconfigurable computing systems. IEEE Rapid Systems Prototyping (1999)
Related Works - II
[*] is the state of art in 3D floorplanning
Simulated Annealing over Transitive Closure Graph
Takes into account device reconfiguration limits
Limits
No heterogeneity considered
High overhead communication
infrastructure solution [**]
12
[*] Ping-Hung Yuh, Chia-Lin Yang, Yao-Wen Chang, Hsin-Lung Chen: Temporal Floorplanning Using 3D-subTCG,
Design Automation Conference, 2004
[**] S. P. Fekete, E. Kohler, and J. Teich: Optimal FPGA Module Placement with Temporal Precedence
Constraints, Proc. DATE, 2001.
FLOORPLACEMENT
13
Floorplanning vs. Placement
Characteristic Floorplanning Placement
# items <100 >10.000
Items (for FPGAs) IP-Core Slice, CLB
Aim Find a position for each item
obj. function depends mainly on Area mainly on Wirelength
Constraints Items can be positioned
everywhere
There is a set of
possible positions
14
Placement
Floor plan
Floorplacement - I
Hierarchical Approach (Floorplanning + Partitioning)
15
S. N. Adya, I. L. Markov, Fixed-outline Floorplanning: Enabling Hierarchical Design, IEEE Transaction on VLSI
System, 2002
S. N. Adya, S. Chaturvedi, J. A. Roy, David A. Papa, I. L. Markov: Unification of Partitioning, Placement and
Floorplanning , IEEE Intl. Conf. on CAD, 2004
Floorplacement - II
Reconfigurable Functional Unit (RFU)
A netlist obtained after post synthesis and technology
mapping (i.e., before placement and routing)
Reconfigurable Region (RR)
16
Floorplacement - III
Resource Aware (i.e., not all positions are feasible)
Device heterogeneity
Device Reconfiguration
capabilities
17
THE PROPOSED APPROACH
18
Proposed Problem Definition
19
Aim
Define RRs
For each task find
find a RR
A position inside RR
Objective Function
Min. Fragmentation
Constraints
Communication issues
Device limits
Target Devices: Xilinx Virtex 4 - 5
Target architecture based on EAPR design flow
Target Architecture and Devices
20
Input
21
Proposed Approach: overview
22
1st Algorithm: Partitioning into RR
Aim: identify the RRs and associate each RFU to one RR
How: partitioning the TG minimizing resource
requirement variance of the RRs (moving and swapping
nodes)
23
Resource of type t
required by RFU n,
at static photo p
2nd Algorithm:TFiRR - I
24
Temporal Floorplacement inside RR (TFiRR)
Aim: for each RR find a set of feasible width-height pairs
How: floorplacing RFUs inside corresponding RR
Assumption: RFUs’ height = height of the RR they belong to
Pseudo Code:
2nd Algorithm:TFiRR - II
25
Let consider an iteration:
2nd Algorithm:TFiRR - III
26
Let consider an iteration:
2nd Algorithm:TFiRR - IV
27
Let consider an iteration:
2nd Algorithm:TFiRR - V
28
Let consider an iteration:
2nd Algorithm:TFiRR - VI
29
2nd Algorithm: TFiRR – Example
30
3rd Algorithm: RR floorplacement - I
Simulated Annealing
Objective Function
Data Structure
4 Constraint Lists (one per row)
31
3rd Algorithm: RR floorplacement - II
Simulated Annealing: moves
Swap two RRs
Move one RRs
Span over one more row
Un-Span over one less row
After each move packing is performed
(i.e., the floorplacement is compressed)
32
3rd Algorithm: RR floorplacement – Example - I
33
3rd Algorithm: RR floorplacement – Example - II
34
Implementation
Three simulated annealers written in C++ STL
35
Output Examples
TFiRR
RR Floorplacement
36
EXPERIMENTAL RESULTS
37
Identification of the number of RRs
38
Partitioning’s impact on TFiRR
TFiRR on
Partitioned TG
TFiRR on TG
Execution Time 125ms 114ms 4m54s
Width
(normalized)
1.00 1.19 1.04
39
Increasing the number of RFUs decreases the possibility
to pick up the right one
Partitioning is a precondition of the 3rd algorithm in order
to better exploit FPGA’s area (2D Floorplacement)
Tests performed directly floorplacing RFUs
Execution time about 100 ms (100K iterations)
Floorplacement – Success Rate
40
Floorplacement – Aspect Ratio
41
Tests performed directly floorplacing RFUs
COMPARISON WITH THE STATE
OF THE ART
42
State of the art
43
Authors Comm.
Infrastructure
Resource
Aware
Reconfiguration
Aware
Device Limits
Aware
Bazargan et al. No No Yes No
Yuh et al. Limited, w/
High Overhead
No Yes Yes
Singhal et al. No No Yes No
Feng et al. No Yes No No
Notes
The comparsion is performed with respect to the
description given by Yuh et al in [*]
Yuh’s approach does not support
Multiple Resources
Existence of a Static side
In order perform the comparison
the case study has been chosen in order to avoid
multiple resource limitation
Yuh’s approach has been extended to support a static
side
44
[*] Ping-Hung Yuh, Chia-Lin Yang, Yao-Wen Chang, Hsin-Lung Chen: Temporal Floorplanning Using 3D-subTCG,
Design Automation Conference, 2004
The Case Study
A Reconfigurable Architecture (for Biomedical Purpose)
on XC5VLX30T
1. Collecting data from sensor
2. Elaborating them
3. Sending to a host computer thorough the net
45
The Proposed Solution
It is a reconfigurable architecture with 2 static photos
46
Area assignments comparison
47
Proposed
solution
Yuh’s solution
Communication performances comparison
Considering 100 M samples, 32 bit each, at 75 MHz.
The entire data are transferred by
The Proposed Approach in 2.6 seconds
Yuh’s approach in 416.0 seconds
48
Conclusions - I
An algorithm for the identification of area constraint
for reconfigurable architectures has been introduced
Novelties: taking into account
Target device heterogeneity
Target device reconfiguration capabilities
Communication issues
49
Conclusions - II
Results have been published
A. Montone, M.D. Santambrogio, D. Sciuto,
A Design Workflow for the Identification of Area
Constraints in Dynamic Reconfigurable Systems,
IEEE International Symposium on Electronic Design, Test
and Applications (DELTA), 2008
A. Montone, M.D. Santambrogio,
Area Constraint Evaluation for FPGAs,
The Syndicated Q1-2008, A technical newsletter for
FPGA, ASIC Verification and DSP Designers,
Synplicity Incorporation
Under revision:
A Reconfiguration-aware Floorplacer for FPGAs,
IEEE Field Programmable Logic (FPL), 2008
50
Future Works
Take into consideration IOBs and inter modules
communications
Partitioning considering clock regions
51
Questions
52

Más contenido relacionado

La actualidad más candente

Andrew Goldberg. Highway Dimension and Provably Efficient Shortest Path Algor...
Andrew Goldberg. Highway Dimension and Provably Efficient Shortest Path Algor...Andrew Goldberg. Highway Dimension and Provably Efficient Shortest Path Algor...
Andrew Goldberg. Highway Dimension and Provably Efficient Shortest Path Algor...
Computer Science Club
 
MIMO System Performance Evaluation for High Data Rate Wireless Networks usin...
MIMO System Performance Evaluation for High Data Rate  Wireless Networks usin...MIMO System Performance Evaluation for High Data Rate  Wireless Networks usin...
MIMO System Performance Evaluation for High Data Rate Wireless Networks usin...
IJMER
 

La actualidad más candente (20)

fft using labview
fft using labviewfft using labview
fft using labview
 
Parallel sorting algorithm
Parallel sorting algorithmParallel sorting algorithm
Parallel sorting algorithm
 
Aw4102359364
Aw4102359364Aw4102359364
Aw4102359364
 
Fft
FftFft
Fft
 
Kernel Descriptors for Visual Recognition
Kernel Descriptors for Visual RecognitionKernel Descriptors for Visual Recognition
Kernel Descriptors for Visual Recognition
 
Transpose Form Fir Filter Design for Fixed and Reconfigurable Coefficients
Transpose Form Fir Filter Design for Fixed and Reconfigurable CoefficientsTranspose Form Fir Filter Design for Fixed and Reconfigurable Coefficients
Transpose Form Fir Filter Design for Fixed and Reconfigurable Coefficients
 
distance_matrix_ch
distance_matrix_chdistance_matrix_ch
distance_matrix_ch
 
Tscd pweb
Tscd pwebTscd pweb
Tscd pweb
 
Vishal Verma: Rapidly Exploring Random Trees
Vishal Verma: Rapidly Exploring Random TreesVishal Verma: Rapidly Exploring Random Trees
Vishal Verma: Rapidly Exploring Random Trees
 
Andrew Goldberg. Highway Dimension and Provably Efficient Shortest Path Algor...
Andrew Goldberg. Highway Dimension and Provably Efficient Shortest Path Algor...Andrew Goldberg. Highway Dimension and Provably Efficient Shortest Path Algor...
Andrew Goldberg. Highway Dimension and Provably Efficient Shortest Path Algor...
 
C42011318
C42011318C42011318
C42011318
 
MIMO System Performance Evaluation for High Data Rate Wireless Networks usin...
MIMO System Performance Evaluation for High Data Rate  Wireless Networks usin...MIMO System Performance Evaluation for High Data Rate  Wireless Networks usin...
MIMO System Performance Evaluation for High Data Rate Wireless Networks usin...
 
Time space trade off
Time space trade offTime space trade off
Time space trade off
 
Space-efficient Feature Maps for String Alignment Kernels
Space-efficient Feature Maps for String Alignment KernelsSpace-efficient Feature Maps for String Alignment Kernels
Space-efficient Feature Maps for String Alignment Kernels
 
Siddhartha fft pptx
Siddhartha fft pptxSiddhartha fft pptx
Siddhartha fft pptx
 
Presentation_Tan
Presentation_TanPresentation_Tan
Presentation_Tan
 
Transformer in Computer Vision
Transformer in Computer VisionTransformer in Computer Vision
Transformer in Computer Vision
 
Mask R-CNN
Mask R-CNNMask R-CNN
Mask R-CNN
 
Hardware Architecture of Complex K-best MIMO Decoder
Hardware Architecture of Complex K-best MIMO DecoderHardware Architecture of Complex K-best MIMO Decoder
Hardware Architecture of Complex K-best MIMO Decoder
 
YolactEdge Review [cdm]
YolactEdge Review [cdm]YolactEdge Review [cdm]
YolactEdge Review [cdm]
 

Destacado (8)

3rd 3DDRESD: BiRF
3rd 3DDRESD: BiRF3rd 3DDRESD: BiRF
3rd 3DDRESD: BiRF
 
Basic Concepts
Basic ConceptsBasic Concepts
Basic Concepts
 
HPPS - Final - 06/14/2007
HPPS - Final - 06/14/2007HPPS - Final - 06/14/2007
HPPS - Final - 06/14/2007
 
UIC Thesis Corbetta
UIC Thesis CorbettaUIC Thesis Corbetta
UIC Thesis Corbetta
 
RCIM 2008 - HLR
RCIM 2008 - HLRRCIM 2008 - HLR
RCIM 2008 - HLR
 
블루오션전략과 창조경영
블루오션전략과 창조경영블루오션전략과 창조경영
블루오션전략과 창조경영
 
RCIM 2008 - - UniCal
RCIM 2008 - - UniCalRCIM 2008 - - UniCal
RCIM 2008 - - UniCal
 
RCIM 2008 - - hArtes Atmel
RCIM 2008 - - hArtes AtmelRCIM 2008 - - hArtes Atmel
RCIM 2008 - - hArtes Atmel
 

Más de Marco Santambrogio (20)

RCIM 2008 - - ALTERA
RCIM 2008 - - ALTERARCIM 2008 - - ALTERA
RCIM 2008 - - ALTERA
 
DHow2 - L6 VHDL
DHow2 - L6 VHDLDHow2 - L6 VHDL
DHow2 - L6 VHDL
 
DHow2 - L6 Ant
DHow2 - L6 AntDHow2 - L6 Ant
DHow2 - L6 Ant
 
DHow2 - L5
DHow2 - L5DHow2 - L5
DHow2 - L5
 
RCIM 2008 - - ALaRI
RCIM 2008 - - ALaRIRCIM 2008 - - ALaRI
RCIM 2008 - - ALaRI
 
RCIM 2008 - Modello Scheduling
RCIM 2008 - Modello SchedulingRCIM 2008 - Modello Scheduling
RCIM 2008 - Modello Scheduling
 
RCIM 2008 -- EHW
RCIM 2008 -- EHWRCIM 2008 -- EHW
RCIM 2008 -- EHW
 
RCIM 2008 - Modello Generale
RCIM 2008 - Modello GeneraleRCIM 2008 - Modello Generale
RCIM 2008 - Modello Generale
 
RCIM 2008 - Allocation Relocation
RCIM 2008 - Allocation RelocationRCIM 2008 - Allocation Relocation
RCIM 2008 - Allocation Relocation
 
RCIM 2008 - - hArtes_Ferrara
RCIM 2008 - - hArtes_FerraraRCIM 2008 - - hArtes_Ferrara
RCIM 2008 - - hArtes_Ferrara
 
RCIM 2008 - Janus
RCIM 2008 - JanusRCIM 2008 - Janus
RCIM 2008 - Janus
 
RCIM 2008 - Intro
RCIM 2008 - IntroRCIM 2008 - Intro
RCIM 2008 - Intro
 
DHow2 - L2
DHow2 - L2DHow2 - L2
DHow2 - L2
 
DHow2 - L4
DHow2 - L4DHow2 - L4
DHow2 - L4
 
DHow2 - L1
DHow2 - L1DHow2 - L1
DHow2 - L1
 
RCW@DEI - Treasure hunt
RCW@DEI - Treasure huntRCW@DEI - Treasure hunt
RCW@DEI - Treasure hunt
 
RCW@DEI - ADL
RCW@DEI - ADLRCW@DEI - ADL
RCW@DEI - ADL
 
RCW@DEI - Design Flow 4 SoPc
RCW@DEI - Design Flow 4 SoPcRCW@DEI - Design Flow 4 SoPc
RCW@DEI - Design Flow 4 SoPc
 
RCW@DEI - Real Needs And Limits
RCW@DEI - Real Needs And LimitsRCW@DEI - Real Needs And Limits
RCW@DEI - Real Needs And Limits
 
RCW@DEI - Basic Concepts
RCW@DEI - Basic ConceptsRCW@DEI - Basic Concepts
RCW@DEI - Basic Concepts
 

Uic Montone Thesis

  • 1. Time-driven reconfiguration-aware floorplacer BY Alessio Montone alessio.montone@dresd.org Thesis committee: S. Dutt (chair), A. Khokhar, M.D. Santambrogio UIC Thesis Defense: 05/08/2008
  • 2. 2 Rationale and Innovation Problem statement Given a reconfigurable architecture, find an on-chip position for each functional unit Innovative contribution: taking into account Target Device Heterogeneity Target Device reconfiguation capabilities Inter-FU Communication
  • 3. 3 Aims Considering the area assignment problem tailored for reconfigurable architectures, provide a formalization of the problem, and an approach (in 3 algorithms) for solving
  • 4. 4 Outline Introduction Floorplacement The Proposed Approach Experimental Results Comparison with the state of the art Conclusions and Future Works Questions
  • 6. Reconfigurable Architectures - I On FPGAs Reconfigurable Devices Heterogeneous Reconfiguration Limits Different types of Reconfigurable Architectures: Total Partial (Static) Partial (Dynamic) 6
  • 8. Reconfigurable Architectures - I Total Partial (Static) 8
  • 9. Reconfigurable Architectures - II Partial Dynamic 9
  • 10. Area Assignment Problem Let consider a Reconfigurable Architecture Given a scheduled task graph (TG) of the application Node: Reconfigurable Functional Unit (RFU) [*], A netlist obtained after post synthesis and technology mapping (i.e., before placement and routing) Aim: find an area assignment for each RFU 10 [*] K. Bazargan, R. Kastner, M.S.: 3-d floorplanning: Simulated annealing and greedy placement methods for reconfigurable computing systems. IEEE Rapid Systems Prototyping (1999)
  • 11. Related Works - I [*] introduced the concept of 3D floorplanning for reconfigurable systems SA in order to solve HW/SW codesign problem For each task choose between HW implementation SW implementation Limits No device limits considered No communication infrastructure 11 [*] K. Bazargan, R. Kastner, M.S.: 3-d floorplanning: Simulated annealing and greedy placement methods for reconfigurable computing systems. IEEE Rapid Systems Prototyping (1999)
  • 12. Related Works - II [*] is the state of art in 3D floorplanning Simulated Annealing over Transitive Closure Graph Takes into account device reconfiguration limits Limits No heterogeneity considered High overhead communication infrastructure solution [**] 12 [*] Ping-Hung Yuh, Chia-Lin Yang, Yao-Wen Chang, Hsin-Lung Chen: Temporal Floorplanning Using 3D-subTCG, Design Automation Conference, 2004 [**] S. P. Fekete, E. Kohler, and J. Teich: Optimal FPGA Module Placement with Temporal Precedence Constraints, Proc. DATE, 2001.
  • 14. Floorplanning vs. Placement Characteristic Floorplanning Placement # items <100 >10.000 Items (for FPGAs) IP-Core Slice, CLB Aim Find a position for each item obj. function depends mainly on Area mainly on Wirelength Constraints Items can be positioned everywhere There is a set of possible positions 14 Placement Floor plan
  • 15. Floorplacement - I Hierarchical Approach (Floorplanning + Partitioning) 15 S. N. Adya, I. L. Markov, Fixed-outline Floorplanning: Enabling Hierarchical Design, IEEE Transaction on VLSI System, 2002 S. N. Adya, S. Chaturvedi, J. A. Roy, David A. Papa, I. L. Markov: Unification of Partitioning, Placement and Floorplanning , IEEE Intl. Conf. on CAD, 2004
  • 16. Floorplacement - II Reconfigurable Functional Unit (RFU) A netlist obtained after post synthesis and technology mapping (i.e., before placement and routing) Reconfigurable Region (RR) 16
  • 17. Floorplacement - III Resource Aware (i.e., not all positions are feasible) Device heterogeneity Device Reconfiguration capabilities 17
  • 19. Proposed Problem Definition 19 Aim Define RRs For each task find find a RR A position inside RR Objective Function Min. Fragmentation Constraints Communication issues Device limits
  • 20. Target Devices: Xilinx Virtex 4 - 5 Target architecture based on EAPR design flow Target Architecture and Devices 20
  • 23. 1st Algorithm: Partitioning into RR Aim: identify the RRs and associate each RFU to one RR How: partitioning the TG minimizing resource requirement variance of the RRs (moving and swapping nodes) 23 Resource of type t required by RFU n, at static photo p
  • 24. 2nd Algorithm:TFiRR - I 24 Temporal Floorplacement inside RR (TFiRR) Aim: for each RR find a set of feasible width-height pairs How: floorplacing RFUs inside corresponding RR Assumption: RFUs’ height = height of the RR they belong to Pseudo Code:
  • 25. 2nd Algorithm:TFiRR - II 25 Let consider an iteration:
  • 26. 2nd Algorithm:TFiRR - III 26 Let consider an iteration:
  • 27. 2nd Algorithm:TFiRR - IV 27 Let consider an iteration:
  • 28. 2nd Algorithm:TFiRR - V 28 Let consider an iteration:
  • 30. 2nd Algorithm: TFiRR – Example 30
  • 31. 3rd Algorithm: RR floorplacement - I Simulated Annealing Objective Function Data Structure 4 Constraint Lists (one per row) 31
  • 32. 3rd Algorithm: RR floorplacement - II Simulated Annealing: moves Swap two RRs Move one RRs Span over one more row Un-Span over one less row After each move packing is performed (i.e., the floorplacement is compressed) 32
  • 33. 3rd Algorithm: RR floorplacement – Example - I 33
  • 34. 3rd Algorithm: RR floorplacement – Example - II 34
  • 38. Identification of the number of RRs 38
  • 39. Partitioning’s impact on TFiRR TFiRR on Partitioned TG TFiRR on TG Execution Time 125ms 114ms 4m54s Width (normalized) 1.00 1.19 1.04 39 Increasing the number of RFUs decreases the possibility to pick up the right one Partitioning is a precondition of the 3rd algorithm in order to better exploit FPGA’s area (2D Floorplacement)
  • 40. Tests performed directly floorplacing RFUs Execution time about 100 ms (100K iterations) Floorplacement – Success Rate 40
  • 41. Floorplacement – Aspect Ratio 41 Tests performed directly floorplacing RFUs
  • 42. COMPARISON WITH THE STATE OF THE ART 42
  • 43. State of the art 43 Authors Comm. Infrastructure Resource Aware Reconfiguration Aware Device Limits Aware Bazargan et al. No No Yes No Yuh et al. Limited, w/ High Overhead No Yes Yes Singhal et al. No No Yes No Feng et al. No Yes No No
  • 44. Notes The comparsion is performed with respect to the description given by Yuh et al in [*] Yuh’s approach does not support Multiple Resources Existence of a Static side In order perform the comparison the case study has been chosen in order to avoid multiple resource limitation Yuh’s approach has been extended to support a static side 44 [*] Ping-Hung Yuh, Chia-Lin Yang, Yao-Wen Chang, Hsin-Lung Chen: Temporal Floorplanning Using 3D-subTCG, Design Automation Conference, 2004
  • 45. The Case Study A Reconfigurable Architecture (for Biomedical Purpose) on XC5VLX30T 1. Collecting data from sensor 2. Elaborating them 3. Sending to a host computer thorough the net 45
  • 46. The Proposed Solution It is a reconfigurable architecture with 2 static photos 46
  • 48. Communication performances comparison Considering 100 M samples, 32 bit each, at 75 MHz. The entire data are transferred by The Proposed Approach in 2.6 seconds Yuh’s approach in 416.0 seconds 48
  • 49. Conclusions - I An algorithm for the identification of area constraint for reconfigurable architectures has been introduced Novelties: taking into account Target device heterogeneity Target device reconfiguration capabilities Communication issues 49
  • 50. Conclusions - II Results have been published A. Montone, M.D. Santambrogio, D. Sciuto, A Design Workflow for the Identification of Area Constraints in Dynamic Reconfigurable Systems, IEEE International Symposium on Electronic Design, Test and Applications (DELTA), 2008 A. Montone, M.D. Santambrogio, Area Constraint Evaluation for FPGAs, The Syndicated Q1-2008, A technical newsletter for FPGA, ASIC Verification and DSP Designers, Synplicity Incorporation Under revision: A Reconfiguration-aware Floorplacer for FPGAs, IEEE Field Programmable Logic (FPL), 2008 50
  • 51. Future Works Take into consideration IOBs and inter modules communications Partitioning considering clock regions 51