darpa i2o transparent computing program
play

DARPA/I2O Transparent Computing Program THEIA: Tagging and Tracking - PowerPoint PPT Presentation

May 29, 2023 •136 likes •1.64k views

DARPA/I2O Transparent Computing Program THEIA: Tagging and Tracking of Multi-Level Host Events for Transparent Computing and Information Assurance Mattia Fazzini Georgia Institute of Technology Nov 3rd, 2017 Agenda Project overview

  1. OS-level Transparency • Goal: – Capture events and dependencies of OS-level events • Approach: – Based on VM introspection • Events analyzed: – Process operations: clone, fork, execve, exit , etc. • – File operations: open, read, write, unlink , etc. • – Network operations: socket, connect, recvmsg , etc. • – Memory operations: mmap, mprotect, shmget , etc . •

  2. OS-level Transparency • Goal: – Capture events and dependencies of OS-level events • Approach: – Based on VM introspection • Events analyzed: – Process operations: clone, fork, execve, exit , etc. • – File operations: open, read, write, unlink , etc. • – Network operations: socket, connect, recvmsg , etc. • – Memory operations: mmap, mprotect, shmget , etc . •

  3. OS-level Transparency • Goal: – Capture events and dependencies of OS-level events • Approach: – Based on VM introspection • Events analyzed: – Process operations: clone, fork, execve, exit , etc. • – File operations: open, read, write, unlink , etc. • – Network operations: socket, connect, recvmsg , etc. • – Memory operations: mmap, mprotect, shmget , etc . •

  4. OS-level Transparency Implementation Example #ifdef TARGET_X86_64 void helper_syscall (int next_eip_addend { panda_cb_list *plist; for(plist = panda_cbs[PANDA_CB_BEFORE_SYSCALL]; plist != NULL; plist = panda_cb_list_next(plist)) { plist->entry.before_syscall(env); } … }

  5. OS-level Transparency Implementation Example #ifdef TARGET_X86_64 void helper_syscall (int next_eip_addend { panda_cb_list *plist; for(plist = panda_cbs[PANDA_CB_BEFORE_SYSCALL]; plist != NULL; plist = panda_cb_list_next(plist)) { plist->entry.before_syscall(env); } … }

  6. OS-level Transparency Implementation Example #ifdef TARGET_X86_64 void helper_syscall (int next_eip_addend { panda_cb_list *plist; for(plist = panda_cbs[PANDA_CB_BEFORE_SYSCALL]; plist != NULL; plist = panda_cb_list_next(plist)) { plist->entry.before_syscall(env); } … }

  7. Action History Graph (AHG) • Goal: – Represent causality across events • Causality: – Process->Process (e.g., fork) – Process->File (e.g., write) – File->Process (e.g., read) – Process->Host (e.g., send) – Host->Process (e.g., recv)

  8. Action History Graph (AHG) • Goal: – Represent causality across events • Causality: – Process->Process (e.g., fork) – Process->File (e.g., write) – File->Process (e.g., read) – Process->Host (e.g., send) – Host->Process (e.g., recv)

  9. Action History Graph (AHG) • Goal: – Represent causality across events • Causality: – Process->Process (e.g., fork) – Process->File (e.g., write) – File->Process (e.g., read) – Process->Host (e.g., send) – Host->Process (e.g., recv)

  10. Action History Graph Example

  11. Coarse-grained Taint Analysis • Goal: – Quickly capture the provenance of objects in the AHG • Working mechanism: – Runs while building AHG – Processes have a provenance set – Process operations: • fork , clone : copy provenance of parent to child process – File and network operations • read , recv : associate provenance of object to process • write , send : associate provenance of process to object

  12. Coarse-grained Taint Analysis • Goal: – Quickly capture the provenance of objects in the AHG • Working mechanism: – Runs while building AHG – Processes have a provenance set – Process operations: • fork , clone : copy provenance of parent to child process – File and network operations • read , recv : associate provenance of object to process • write , send : associate provenance of process to object

  13. Coarse-grained Taint Analysis • Goal: – Quickly capture the provenance of objects in the AHG • Working mechanism: – Runs while building AHG – Processes have a provenance set – Process operations: • fork , clone : copy provenance of parent to child process – File and network operations • read , recv : associate provenance of object to process • write , send : associate provenance of process to object

  14. Fine-grained Taint Analysis • Goal: – Accurately capture provenance of objects in the AHG • Working mechanism: – Decoupled from program execution – Instruction level propagation – Taint tags at byte level granularity • Optimizations: – Trace-based dynamic taint analysis

  15. Fine-grained Taint Analysis • Goal: – Accurately capture provenance of objects in the AHG • Working mechanism: – Decoupled from program execution – Instruction level propagation – Taint tags at byte level granularity • Optimizations: – Trace-based dynamic taint analysis

  16. Fine-grained Taint Analysis • Goal: – Accurately capture provenance of objects in the AHG • Working mechanism: – Decoupled from program execution – Instruction level propagation – Taint tags at byte level granularity • Optimizations: – Trace-based dynamic taint analysis

  17. Fine-grained Taint Analysis Implementation TCG LLVM Guest Basic Block Basic Block Basic Block

  18. Fine-grained Taint Analysis Implementation TCG LLVM Guest Basic Block Basic Block Basic Block

  19. Fine-grained Taint Analysis Implementation TCG LLVM Guest Basic Block Basic Block Basic Block

  20. Fine-grained Taint Analysis Implementation TCG LLVM Guest Basic Block Basic Block Basic Block

  21. Trace-based Taint Analysis • Objective: – Improve performance of fine-grained taint analysis • Key intuition: – Within a trace instruction sequences are executed multiple times • Working mechanism: – Based on the execution trace of the system/program – Computes taint summaries for sequences of instructions – Re-use taint summaries on the trace and possible across traces • Implementation: – Sequitur algorithm: recognizes a lexical structure in an execution trace and generates a grammar where terminals are instructions – Analyze grammar and reuse taint results when possible

  22. Trace-based Taint Analysis • Objective: – Improve performance of fine-grained taint analysis • Key intuition: – Within a trace instruction sequences are executed multiple times • Working mechanism: – Based on the execution trace of the system/program – Computes taint summaries for sequences of instructions – Re-use taint summaries on the trace and possible across traces • Implementation: – Sequitur algorithm: recognizes a lexical structure in an execution trace and generates a grammar where terminals are instructions – Analyze grammar and reuse taint results when possible

  23. Trace-based Taint Analysis • Objective: – Improve performance of fine-grained taint analysis • Key intuition: – Within a trace instruction sequences are executed multiple times • Working mechanism: – Based on the execution trace of the system/program – Computes taint summaries for sequences of instructions – Re-use taint summaries on the trace and possible across traces • Implementation: – Sequitur algorithm: recognizes a lexical structure in an execution trace and generates a grammar where terminals are instructions – Analyze grammar and reuse taint results when possible

  24. Trace-based Taint Analysis Example Execution Trace Grammar mov qword ptr [r12+rax*8], rdx 9 10 jump 0x7f8c47a21b13 … mov qword ptr [r12+rax*8], rdx jz 0x7f8c47a21b52 jmp 0x7f8c47a21b13 476 add rdx, 0x10 11 11 8 mov rax, qword ptr [rdx] test rax, rax add rdx, 0x10 jz 0x7f8c47a21b52 43 cmp rax, 0x21 jbe 0x7f8c47a21b08 lea rcx, ptr [rip+0x21ef29] … mov rax, qword ptr [rdx] test rax, rax

  25. Trace-based Taint Analysis Example Execution Trace Grammar mov qword ptr [r12+rax*8], rdx 9 10 jump 0x7f8c47a21b13 … mov qword ptr [r12+rax*8], rdx jz 0x7f8c47a21b52 jmp 0x7f8c47a21b13 476 add rdx, 0x10 11 11 8 mov rax, qword ptr [rdx] test rax, rax add rdx, 0x10 jz 0x7f8c47a21b52 43 cmp rax, 0x21 jbe 0x7f8c47a21b08 lea rcx, ptr [rip+0x21ef29] … mov rax, qword ptr [rdx]

  26. Fine-grained Taint Analysis

  27. Fine-grained Taint Analysis

  28. Case Study Overview

  29. Case Study Overview

  30. Case Study Overview

  31. Case Study Overview

  32. Case Study Overview

  33. Case Study and AHG bash execute firefox Process recv firefox from Event 143.215.130.204 File execute sh Network wget 143.215.130.204 sh Tag execute recv wget from Causality screen execute write grab recv screen msg grab X0 write s.png nc execute nc read write 143.215.130.204

  34. Case Study and AHG bash execute firefox Process recv firefox from Event 143.215.130.204 File execute sh Network wget 143.215.130.204 sh Tag execute recv wget from Causality screen execute write grab recv screen msg grab X0 write s.png nc execute nc read write 143.215.130.204

  35. Case Study and AHG Step 1 1) Victim starts Firefox Process Event bash File Network Tag execute firefox recv from recv from screen grab firefox recv screen msg grab

  36. Case Study and AHG Step 2 2) Victim visits malicious.com (143.215.130.204) that runs shell process Process Event recv firefox from File Network 143.215.130.204 Tag execute sh recv from recv sh from screen grab recv screen msg grab

  37. Case Study and AHG Step 3 3) Attacker downloads and executes screengrab Process wget Event 143.215.130.204 sh execute File recv wget Network from Tag screen execute write grab recv from recv screen X0 recv msg grab from screen grab recv screen msg grab write s.png

  38. Case Study and AHG Step 4 4) Screenshot is sent to attacker’s server Process nc Event sh execute File Network s.png nc read Tag write recv from 143.215.130.204 recv from screen grab recv screen msg grab

  39. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read Process Event libc.so read File sh execute Network Tag Coarse Taint Set wgetrc read wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  40. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read File sh execute Network Tag Coarse Taint Set wgetrc read wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  41. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read File sh execute Network Tag Coarse Taint Set wgetrc read CT1 wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  42. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read CT2 File sh execute Network Tag Coarse Taint Set wgetrc read CT1 wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  43. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read CT2 File sh execute Network Tag Coarse Taint Set wgetrc read CT1 CT2 wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  44. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read CT2 File sh execute Network Tag Coarse Taint Set wgetrc read CT3 CT1 CT2 wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  45. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read CT2 File sh execute Network Tag Coarse Taint Set wgetrc read CT3 CT1 CT2 CT3 wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  46. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read CT2 File sh execute Network Tag Coarse Taint Set wgetrc read CT3 CT1 CT2 CT3 wget recv from recv CT4 from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  47. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read CT2 File sh execute Network Tag Coarse Taint Set wgetrc read CT3 CT1 CT2 CT3 CT4 wget recv from recv CT4 from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  48. Case Study and 
 Coarse-grained Taint Analysis. libssl.so wget read CT1 Process Event libc.so read CT2 File sh execute Network Tag Coarse Taint Set wgetrc read CT3 CT1 CT2 CT3 CT4 wget recv from recv CT4 from recv from 143.215.130.204 screen grab recv screen msg grab screen write CT5 grab

  49. Case Study and 
 Fine-grained Taint Analysis libssl.so wget read Process Event libc.so read File sh execute Network Tag wgetrc read wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  50. Case Study and 
 Fine-grained Taint Analysis libssl.so wget read FT1 Process Event libc.so read File sh execute Network Tag wgetrc read wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  51. Case Study and 
 Fine-grained Taint Analysis libssl.so wget read FT1 Process Event libc.so read FT2 File sh execute Network Tag wgetrc read wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  52. Case Study and 
 Fine-grained Taint Analysis libssl.so wget read FT1 Process Event libc.so read FT2 File sh execute Network Tag wgetrc read FT3 wget recv from recv from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  53. Case Study and 
 Fine-grained Taint Analysis libssl.so wget read FT1 Process Event libc.so read FT2 File sh execute Network Tag wgetrc read FT3 wget recv from recv FT4 from recv from 143.215.130.204 screen grab recv screen msg grab screen write grab

  54. Case Study and 
 Fine-grained Taint Analysis libssl.so wget read FT1 Process Event libc.so read FT2 File sh execute Network Tag wgetrc read FT3 wget recv from recv FT4 from recv from 143.215.130.204 screen grab recv screen msg grab screen write FT5 grab

  55. THEIA-Panda Overheads Bare Exec KVM Exec QEMU Record Replay TIME Time Time Exec Time Exec Time Exec Time Bare Exec Time KVM Exec 2.09 x Time QEMU 6.19 x 2.96 x Exec Time Record 7.75 x 3.71 x 1.25 x Exec Time Replay 13.82 x 6.62 x 2.23 x 1.78 x Exec Time • Fine grained taint analysis: – ~40x to ~300x compared to bare execution • Space overhead: – ~86 GB/day non det log data + ~1.3GB/day graph data

  56. THEIA-Panda Overheads Bare Exec KVM Exec QEMU Record Replay TIME Time Time Exec Time Exec Time Exec Time Bare Exec Time KVM Exec 2.09 x Time QEMU 6.19 x 2.96 x Exec Time Record 7.75 x 3.71 x 1.25 x Exec Time Replay 13.82 x 6.62 x 2.23 x 1.78 x Exec Time • Fine grained taint analysis: – ~40x to ~300x compared to bare execution • Space overhead: – ~86 GB/day non det log data + ~1.3GB/day graph data

  57. THEIA-Panda Overheads Bare Exec KVM Exec QEMU Record Replay TIME Time Time Exec Time Exec Time Exec Time Bare Exec Time KVM Exec 2.09 x Time QEMU 6.19 x 2.96 x Exec Time Record 7.75 x 3.71 x 1.25 x Exec Time Replay 13.82 x 6.62 x 2.23 x 1.78 x Exec Time • Fine grained taint analysis: – ~40x to ~300x compared to bare execution • Space overhead: – ~86 GB/day non det log data + ~1.3GB/day graph data

  58. THEIA-Panda Overheads Bare Exec KVM Exec QEMU Record Replay TIME Time Time Exec Time Exec Time Exec Time Bare Exec Time KVM Exec 2.09 x Time QEMU 6.19 x 2.96 x Exec Time Record 7.75 x 3.71 x 1.25 x Exec Time Replay 13.82 x 6.62 x 2.23 x 1.78 x Exec Time • Fine grained taint analysis: – ~40x to ~300x compared to bare execution • Space overhead: – ~86 GB/day non det log data + ~1.3GB/day graph data

  59. THEIA-Panda Overheads Bare Exec KVM Exec QEMU Record Replay TIME Time Time Exec Time Exec Time Exec Time Bare Exec Time KVM Exec 2.09 x Time QEMU 6.19 x 2.96 x Exec Time Record 7.75 x 3.71 x 1.25 x Exec Time Replay 13.82 x 6.62 x 2.23 x 1.78 x Exec Time • Fine grained taint analysis: – ~40x to ~300x compared to bare execution • Space overhead: – ~86 GB/day non det log data + ~1.3GB/day graph data

  60. THEIA-Panda Observations -Panda

  61. THEIA-KI Overview THEIA-KI Query Interface + OS Replay FA Record Fine-grained Storage Taint Analysis System Call Information Action History Graph Process Information Real-time THEIA-KI-Analysis On-demand

Recommend

Appendix 2 - Textual Analysis Rubric Thematic comparison Indigenous Categories Key Words in context

Appendix 2 - Textual Analysis Rubric Thematic comparison Indigenous Categories Key Words in context

DARPA Appendix 1 - DARPA Site Map Quality Hiring Privacy & Budget Doing Director's Grand News Mission & DARPA Program Challenge Security of Info. Office Releases Info. Programs Business Overview Managers with DARPA

557 views • 9 slides
AADL for DoD Systems Dr. Raymond Richards Program Manager DARPA/I2O AADL Users Days

AADL for DoD Systems Dr. Raymond Richards Program Manager DARPA/I2O AADL Users Days

AADL for DoD Systems Dr. Raymond Richards Program Manager DARPA/I2O AADL Users Days November 14, 2019 Distribution A. Approved for public release: distribution unlimited. DARPA Efforts in Digital Engineering High Assurance Cyber Military

335 views • 9 slides
MD.MFP Dual Use Technology Briefing Larry Herriman, Assistant Executive Director of the

MD.MFP Dual Use Technology Briefing Larry Herriman, Assistant Executive Director of the

MD.MFP Dual Use Technology Briefing Larry Herriman, Assistant Executive Director of the Macomb-OU INCubator Director of the Michigan DARPA Matching Funds Program April 10, 2013 MD.MFP DARPA Organizational Overview 2 MD.MFP What is DARPA?

225 views • 21 slides
DARPAs I nvestments in Real World Cryptography Dr Josh Baron Program Manager, DARPA/I2O Real

DARPAs I nvestments in Real World Cryptography Dr Josh Baron Program Manager, DARPA/I2O Real

DARPAs I nvestments in Real World Cryptography Dr Josh Baron Program Manager, DARPA/I2O Real World Cryptography 2019 9 January 2019 1 Distribution Statement A: Approved for Public Release, Distribution Unlimited Cryptography at DARPA

324 views • 21 slides
Other Transactions Agreements (OTs): A Tool for Enhancing CSRE-DARPA Collaboration Crane Lopes

Other Transactions Agreements (OTs): A Tool for Enhancing CSRE-DARPA Collaboration Crane Lopes

Other Transactions Agreements (OTs): A Tool for Enhancing CSRE-DARPA Collaboration Crane Lopes General Counsel Defense Advanced Research Projects Agency (DARPA) https://www.darpa.mil/ January 15, 2019 Distribution Statement A 1 Bottom Line

254 views • 22 slides
Transparent migration of virtual Transparent migration of virtual infrastructures in large

Transparent migration of virtual Transparent migration of virtual infrastructures in large

Transparent migration of virtual Transparent migration of virtual infrastructures in large datacenters for infrastructures in large datacenters for Cloud Computing Cloud Computing Workshop on Management of Cloud Systems ( Workshop on

489 views • 23 slides
Silicon Compilers - Version 2.0 Andreas Olofsson Program Manager, DARPA/MTO I nternational

Silicon Compilers - Version 2.0 Andreas Olofsson Program Manager, DARPA/MTO I nternational

Silicon Compilers - Version 2.0 Andreas Olofsson Program Manager, DARPA/MTO I nternational Symposium on Physical Design March 25-28, Monterey, CA Distribution Statement A (Approved for Public Release, Distribution Unlimited) A Brief

642 views • 42 slides
DARPA HPCS Overview Productivity Evaluation David Koester, Ph.D. DARPA HPCS Productivity Team

DARPA HPCS Overview Productivity Evaluation David Koester, Ph.D. DARPA HPCS Productivity Team

DARPA HPCS Overview Productivity Evaluation David Koester, Ph.D. DARPA HPCS Productivity Team HPCchallenge Benchmarks Panel SC2004 12 November 2004 This work is sponsored by the Department of Defense under Army Contract

378 views • 11 slides
The Role of DARPA Ed Lazowska History of Computing Autumn 2006 1 Overview of Tire Tracks

The Role of DARPA Ed Lazowska History of Computing Autumn 2006 1 Overview of Tire Tracks

The Role of DARPA Ed Lazowska History of Computing Autumn 2006 1 Overview of Tire Tracks Diagram Shows 19 $1B (or larger) sub-sectors of IT Shows university research (federal funding), industry research (industry or federal

1.02k views • 33 slides
Making spatial aggregation more transparent Amelia McNamara @AmeliaMN Currently: Program in

Making spatial aggregation more transparent Amelia McNamara @AmeliaMN Currently: Program in

Making spatial aggregation more transparent Amelia McNamara @AmeliaMN Currently: Program in Statistical and Data Sciences, Smith College, Northampton MA Fall 2018: Department of Computer & Information Sciences, University of St Thomas, St

612 views • 8 slides
Transparent Assessment Providing transparent goals and expectations for students Jonathon Adams

Transparent Assessment Providing transparent goals and expectations for students Jonathon Adams

Transparent Assessment Providing transparent goals and expectations for students Jonathon Adams Shinshu University Thursday, 9 August 12 Outline Aims of the presentation An approach to transparent goals and expectations Example

576 views • 46 slides
Application-Transparent Checkpoint/Restart for MPI Programs over InfiniBand Qi Gao, Weikuan Yu,

Application-Transparent Checkpoint/Restart for MPI Programs over InfiniBand Qi Gao, Weikuan Yu,

Application-Transparent Checkpoint/Restart for MPI Programs over InfiniBand Qi Gao, Weikuan Yu, Wei Huang, Dhabaleswar K. Panda Network-Based Computing Laboratory Department of Computer Science & Engineering The Ohio State University

635 views • 26 slides
Device-Transparent Personal Storage Based on the article Eyo: Device Transparent Personal

Device-Transparent Personal Storage Based on the article Eyo: Device Transparent Personal

Device-Transparent Personal Storage Based on the article Eyo: Device Transparent Personal Storage by Jacob Strauss, Justin Mazzola Paluska, Chris Lesniewski-Laas, Bryan Ford, Robert Morris, Frans Kaashoek Eyos assumptions and API

490 views • 30 slides
Simulating Transparent Migration in Java Java doesnt provide transparent migration. non

Simulating Transparent Migration in Java Java doesnt provide transparent migration. non

Simulating Transparent Migration in Java Java doesnt provide transparent migration. non transparent programm as if mymethod(int x) mymethod() { { if ( x==1) Use code instrumentation go() go() if ( x==2) } } save local

322 views • 13 slides
Gazebo: Simulator for the DARPA VRC Steven Peters, John Hsu gazebosim.org Outline Overview of

Gazebo: Simulator for the DARPA VRC Steven Peters, John Hsu gazebosim.org Outline Overview of

Gazebo: Simulator for the DARPA VRC Steven Peters, John Hsu gazebosim.org Outline Overview of the Open Source Gazebo Simulator DARPA Virtual Robotics Challenge Challenges with real-time simulation Open Dynamics Engine (ODE) Projected

757 views • 30 slides
Transparent boundary conditions for the elastic Transparent boundary conditions for the elastic

Transparent boundary conditions for the elastic Transparent boundary conditions for the elastic

Transparent boundary conditions for the elastic Transparent boundary conditions for the elastic waves in anisotropic media waves in anisotropic media Ivan L. Sofronov, Nickolai A. Zaitsev Keldysh Institute of Applied Mathematics RAS, Moscow

514 views • 35 slides
FTRMI: Fault-Tolerant Transparent RMI Diogo Reis and Hugo Miranda University of Lisbon LaSIGE

FTRMI: Fault-Tolerant Transparent RMI Diogo Reis and Hugo Miranda University of Lisbon LaSIGE

FTRMI: Fault-Tolerant Transparent RMI Diogo Reis and Hugo Miranda University of Lisbon LaSIGE SAC, March 28th, 2012 Remote Procedure Calls What? A simple approach for distributed computing Hides the network from the application (client and

348 views • 20 slides
All, I have drafted a template for us to use to create the DARPA STO slides. Recall that these

All, I have drafted a template for us to use to create the DARPA STO slides. Recall that these

From: John Mullet To: Bill Rooney; Stelly_David Stelly; Adam Helms; Bob Avant ; Steve Searcy; Bill McCutchen Subject: Draft DARPA STO slides Date: Monday, October 12, 2009 11:42:53 AM Attachments: DARPA_STO slides_081209.pptx ATT00007.txt

361 views • 15 slides
Using Friendly Tail Bounds for Sums of Random Matrices Joel A. Tropp Computing +

Using Friendly Tail Bounds for Sums of Random Matrices Joel A. Tropp Computing +

Using Friendly Tail Bounds for Sums of Random Matrices Joel A. Tropp Computing + Mathematical Sciences California Institute of Technology jtropp@cms.caltech.edu Research supported in part by NSF, DARPA, ONR, and AFOSR 1 . Matrix .

726 views • 26 slides
CompSci 514: Computer Networks Lecture 3: The Design Philosophy of the DARPA Internet Protocols

CompSci 514: Computer Networks Lecture 3: The Design Philosophy of the DARPA Internet Protocols

CompSci 514: Computer Networks Lecture 3: The Design Philosophy of the DARPA Internet Protocols Xiaowei Yang xwy@cs.duke.edu Overview The Design Philosophy of the DARPA Internet Protocols What are the important goals How prioritizing

527 views • 37 slides
HPC Strategy & US Exascale Program James Amundson, Scientific Computing Division Head

HPC Strategy & US Exascale Program James Amundson, Scientific Computing Division Head

HPC Strategy & US Exascale Program James Amundson, Scientific Computing Division Head Inaugural Meeting of the International Computing Advisory Committee March 14, 2019 My Perspective The vast majority of HEP computing to date has been

717 views • 13 slides
Introduction 1989: REAL network simulator Ns Tutorial 2002 1995: DARPA VINT project at

Introduction 1989: REAL network simulator Ns Tutorial 2002 1995: DARPA VINT project at

Introduction 1989: REAL network simulator Ns Tutorial 2002 1995: DARPA VINT project at LBL, Xerox PARC, UCB, and USC/ISI Present: DARPA SAMAN project and Padmaparna Haldar (haldar@isi.edu) NSF CONSER project Xuan Chen

584 views • 9 slides
User-Friendly Tools for Random Matrices Joel A. Tropp Computing + Mathematical Sciences

User-Friendly Tools for Random Matrices Joel A. Tropp Computing + Mathematical Sciences

User-Friendly Tools for Random Matrices Joel A. Tropp Computing + Mathematical Sciences California Institute of Technology jtropp@cms.caltech.edu Research supported by ONR, AFOSR, NSF, DARPA, Sloan, and Moore. 1 . Download the Notes:

489 views • 48 slides
Program Families in Scientific Computing Methodology Mesh Generator Generator Virtual Spencer

Program Families in Scientific Computing Methodology Mesh Generator Generator Virtual Spencer

Slide 1 of 20 Why Program Families? Program Families in Scientific Computing Methodology Mesh Generator Generator Virtual Spencer Smith, John McCutchan and Fang Cao Laboratories Concluding Remarks Department of Computing and Software

409 views • 20 slides

More recommend