PCaPAC2014 - List of Keywords (distributed)

Paper	Title	Other Keywords	Page
WPO029	Implementation of the Distributed Alarm System for the Particle Accelerator FAIR Using an Actor Concurrent Programming Model and the Concept of an Agent	framework, monitoring, software, simulation	102
	D. Kumar, G.G. Gaperic, M. Pleško Cosylab, Ljubljana, Slovenia R. Huhmann, S. Krepp GSI, Darmstadt, Germany
	The Alarm System is a software system that enables operators to identify and locate conditions which indicate hardware and software components malfunctioning or nearby malfunctioning. The FAIR Alarm System is being constructed as a Slovenian in-kind contribution to FAIR project. The purpose of the paper is to show how to simplify the development of a highly available distributed alarm system for the particle accelerator FAIR using a concurrent programming model based on actors and on the concept of an agent. The agents separate the distribution of the alarm status signals to the clients from the processing of the alarm signals. The logical communication between an alarm client and an agent is between an actor in the alarm client and an actor in the agent. These two remote actors exchange messages through Java MOM. The following will be addressed: the tree-like hierarchy of actors that are used for the fault tolerance communication between an agent and an alarm client; a custom message protocol used by the actors; the message system and corresponding technical implications; and details of software components that were developed using the Akka programming library.

TCO301	Inexpensive Scheduling in FPGAs	FPGA, hardware, controls, interface	150
	W.W. Terpstra, D.H. Beck, M. Kreider GSI, Darmstadt, Germany
	In the new scheme for machine control used within the FAIR project, actions are distributed to front-end controllers (FEC) with absolute execution timestamps. The execution time must be both precise to the nanosecond and scheduled faster than a microsecond, requiring a hardware solution. Although the actions are scheduled at the FEC out of order, they must be executed in sorted order. The typical hardware approaches to implementing a priority queue (CAMs, shift-registers, etc.) work well in ASIC designs, but must be implemented in expensive FPGA core logic. Conversely, the typical software approaches (heaps, calendar queues, etc.) are either too slow or too memory intensive. We present an approach which exploits the time-ordered nature of our problem to sort in constant-time using only a few memory blocks.
	Slides TCO301 [1.370 MB]

FPO016	Status of Operation Data Archiving System Using Hadoop/HBase for J-PARC	database, operation, status, EPICS	196
	N. Kikuzawa, Y. Kato, A. Yoshii JAEA/J-PARC, Tokai-Mura, Naka-Gun, Ibaraki-Ken, Japan H. Ikeda, N. Ouchi JAEA, Ibaraki-ken, Japan
	J-PARC (Japan Proton Accelerator Research Complex) consists of much equipment. In Linac and 3 GeV rapid cycling synchrotron ring (RCS), the data of over the 64,000 EPICS records for these equipment has been collected. The Data volume is about 2 TB in every year, and the stored total data volume is about 10 TB. The data have been being stored by a Relational Data Base (RDB) system using PostgreSQL, but it is not enough in availability, performance, and capability to increase of data volume flexibility. Hadoop/HBase, which is known as a distributed, scalable and big data store, has been proposed for our next-generation archive system to solve these problems. The test system was built and verified about data transition or database utilization. This report shows the current status of the new archive system, and its advantages and problems which have been obtained through our verification.

Paper

Title

Other Keywords

Page

Implementation of the Distributed Alarm System for the Particle Accelerator FAIR Using an Actor Concurrent Programming Model and the Concept of an Agent

framework, monitoring, software, simulation

102

D. Kumar, G.G. Gaperic, M. Pleško
Cosylab, Ljubljana, Slovenia
R. Huhmann, S. Krepp
GSI, Darmstadt, Germany

The Alarm System is a software system that enables operators to identify and locate conditions which indicate hardware and software components malfunctioning or nearby malfunctioning. The FAIR Alarm System is being constructed as a Slovenian in-kind contribution to FAIR project. The purpose of the paper is to show how to simplify the development of a highly available distributed alarm system for the particle accelerator FAIR using a concurrent programming model based on actors and on the concept of an agent. The agents separate the distribution of the alarm status signals to the clients from the processing of the alarm signals. The logical communication between an alarm client and an agent is between an actor in the alarm client and an actor in the agent. These two remote actors exchange messages through Java MOM. The following will be addressed: the tree-like hierarchy of actors that are used for the fault tolerance communication between an agent and an alarm client; a custom message protocol used by the actors; the message system and corresponding technical implications; and details of software components that were developed using the Akka programming library.

TCO301

Inexpensive Scheduling in FPGAs

FPGA, hardware, controls, interface

150

W.W. Terpstra, D.H. Beck, M. Kreider
GSI, Darmstadt, Germany

In the new scheme for machine control used within the FAIR project, actions are distributed to front-end controllers (FEC) with absolute execution timestamps. The execution time must be both precise to the nanosecond and scheduled faster than a microsecond, requiring a hardware solution. Although the actions are scheduled at the FEC out of order, they must be executed in sorted order. The typical hardware approaches to implementing a priority queue (CAMs, shift-registers, etc.) work well in ASIC designs, but must be implemented in expensive FPGA core logic. Conversely, the typical software approaches (heaps, calendar queues, etc.) are either too slow or too memory intensive. We present an approach which exploits the time-ordered nature of our problem to sort in constant-time using only a few memory blocks.

Slides TCO301 [1.370 MB]

FPO016

Status of Operation Data Archiving System Using Hadoop/HBase for J-PARC

database, operation, status, EPICS

196

N. Kikuzawa, Y. Kato, A. Yoshii
JAEA/J-PARC, Tokai-Mura, Naka-Gun, Ibaraki-Ken, Japan
H. Ikeda, N. Ouchi
JAEA, Ibaraki-ken, Japan

J-PARC (Japan Proton Accelerator Research Complex) consists of much equipment. In Linac and 3 GeV rapid cycling synchrotron ring (RCS), the data of over the 64,000 EPICS records for these equipment has been collected. The Data volume is about 2 TB in every year, and the stored total data volume is about 10 TB. The data have been being stored by a Relational Data Base (RDB) system using PostgreSQL, but it is not enough in availability, performance, and capability to increase of data volume flexibility. Hadoop/HBase, which is known as a distributed, scalable and big data store, has been proposed for our next-generation archive system to solve these problems. The test system was built and verified about data transition or database utilization. This report shows the current status of the new archive system, and its advantages and problems which have been obtained through our verification.