11
Presented by Logistical networking for energy sciences Micah Beck Computer Science and Mathematics Computing and Computational Sciences Directorate Logistical Computing and Internetworking (LoCI) Lab Computer Science Department University of Tennessee

Logistical networking for energy sciences

  • Upload
    caroun

  • View
    24

  • Download
    1

Embed Size (px)

DESCRIPTION

Logistical networking for energy sciences. Micah Beck Computer Science and Mathematics Computing and Computational Sciences Directorate Logistical Computing and Internetworking (LoCI) Lab Computer Science Department University of Tennessee. The Internet backplane protocol (IBP). - PowerPoint PPT Presentation

Citation preview

Page 1: Logistical networking  for energy sciences

Presented by

Logistical networking for energy sciences

Micah BeckComputer Science and Mathematics

Computing and Computational Sciences Directorate

Logistical Computing and Internetworking (LoCI) Lab Computer Science Department

University of Tennessee

Page 2: Logistical networking  for energy sciences

2 Beck_LNES_0611

The Internet backplane protocol (IBP) IBP is a common service for state

management in a shared network

It is a basis for asynchronous communication

Scalability comes from weak assumptions: Maximum size and duration of allocation A highly generic, “best effort” service “A weak network version of malloc”

Robust services are built on top in an end-to-end manner !

The goal is scalability analogous to the Internet

2 Beck_LNES_0611

Page 3: Logistical networking  for energy sciences

3 Beck_LNES_0611

Internet backplane protocol

NaNa

Allocate!

Capability

Depot

NwNw

Store!

DepotData

NrNr

Load!

Depot

Page 4: Logistical networking  for energy sciences

4 Beck_LNES_0611

Internet backplane protocol

NaNa

Copy!Capability

Depot

Depot

NrNr

Compute!

Depot

Page 5: Logistical networking  for energy sciences

5 Beck_LNES_0611

Apphost

LoDN/L-Store

Three kinds of files Files stored locally (1)

Attached disk (direct, LAN or storage network)

Files represented as exNodes exNodes stored locally (2) exNodes stored in LoDN directory (3)

Page 6: Logistical networking  for energy sciences

6 Beck_LNES_0611

One infrastructure serves all

LN provides a “bits are bits” infrastructure

Standardize on what we have an adequate common model for: Storage/buffer management Coarse-grained data transfer

Leave everything else to higher layers End-to-end services: checksums,

encryption, error encoding, etc.

Enable autonomy in wide area service creation: security, resource allocation, QoS guarantees…

Gain the benefits of interoperability today

Page 7: Logistical networking  for energy sciences

7 Beck_LNES_0611

EwokclusterEwok

cluster

Fusion: Content distribution

Depots

JaguarCray XT3Jaguar

Cray XT3

NYU

PPPL

UCI

MIT

Portals

Directoryserver

Page 8: Logistical networking  for energy sciences

8 Beck_LNES_0611

NYU

PPPL

UCI

MIT

end user

Fusion: Location Independent I/0

Retrievemetadata

Retrieve data from best-performing depots

Directoryserver

8 Beck_LNES_0611

2.

1.

Page 9: Logistical networking  for energy sciences

9 Beck_LNES_0611

SRM-Interoperability

Page 10: Logistical networking  for energy sciences

10 Beck_LNES_0611

Page 11: Logistical networking  for energy sciences

11 Beck_LNES_0611

Contact

Micah BeckLogistical Computing and Internetworking (LoCI) Lab Computer Science Department University of Tennessee(865) [email protected]

11 Beck_LNES_0611