More information regarding the bug]]>
more information at: Supporting Domain Specific Programming]]>
This is a paper that outline the needs smart controlling and monitoring of small electricity generators:
And a few ( 3 ) years later:
Future Smart Grid in USA
Btw, it is work in progress]]>
The picture came from the Amazon developer guide and is more or less a good summary. In few words you pay for CPUs, data coming in and out amazon and permanent storage.
What is not clearly shown in the picture is the costs for the amazon services. In other words every time you perform a change or monitor in the infrastructure with external tools you must pay something because you are using the web services of amazon. If you have a lot of datas or you are using many virtual machine this cost is negligible but if you plan a close monitoring to your infrastructure you should be prepare to add this extra costs.]]>
Different possibilities have been explored to cope with this problem in diverse scenarios in which machines and Instruments are deployed. This kind of demand becomes even more crucial in those cases characterized by a huge number of sensors/probes involving a highly dynamic change in their distribution and inter-connections.
Peer to peer has been proposed as a possible approach for covering the aforementioned need. With this solution different instruments can discover information by the others and cooperate to optimize the overall system performances and/or repair possible system faults.
This video presents a simple and intuitive demonstration of self-configuration and self-optimization properties in a set of Instruments. Here you have the possibility to see the system deployment and to trigger a re-configuration by adding a new Instrument.
The map of this video shows several different markers, each representing one or more instruments spread about the entire world.
Different colours stand for different types of Instruments: adding a new marker on the map, that means to join your machine, as a new Instrument, into the Instrument Network, is as easy as clicking a “Join!” button.
In the Demo you will see the following actions:
1) Server that hosts a geo-location of existing devices is displayed
2) A sensor application is started in a local machine using javaWeb start.
3) The Map is automatically updated with the location and the information of the new node.
4) The sensor is de-instantiated and it disappear from the map.
5) The basic GUI of the core machine that maintains the information of the index is displayed
6) Different ways of showing the information acquired from the sensors are displayed
2 different ways to deal with the low level communications have been provided in order to support an access from a machine where the ARC-User Interface is not installed. This is he case of a remote user interface and he APIs support an SSH connection with the machine where the ARC command line interface runs in order to read/write strings into the wire and performs file transfers (using SCP).
Grid security is supported via an automatic regeneration of the user certificate and is optional. In other words, if you can assume that the machine where the User Interface is running support has a valid proxy installed the security become optional and can be turned off.
The key class is ARCFacade.java that Abstract the usual job submission functionalities and gives a “method oriented” way to interact with ARC.
Example are: killjob, renewCertificates, submitjob etc etc.
In the release an example of usage is provided. And you can have the following information:
1) Validate the configuration,
2) Connect to the remote user interface,
3) Upload few files into the remote machine
4) Submit a job,
5) Wait (monitor the status of the job)
6) Retrieve the results from the grid
7) Move the results from the remote machine to the machine where the jARC is running
For more details you can refer to:
The Swiss Grid School 2010 (SGS’10) is organized by the Swiss National Grid (SwiNG) association. Created in October 2007, SwiNG promotes Grid computing in several scientific and industry-related domains (physics, chemistry, life science, engineering, finance, computer science, etc.). One of SwiNG’s objectives is to run education and outreach activities. In this context, SGS’10 aims at transferring Grid knowledge to academia, research and industry. The following audience is particularly addressed:
SGS’10 provides lectures and practical lab sessions that illustrate the current state-of-the-art in Grid computing in the following domains: Grid architecture, security and middleware, resource management, data management and scientific workflow management. It also focuses on a variety of practical case studies (applications).]]>
As we can see from this talk the USA government has a growing interest in the Smart Grid topic:
Future Smart Grid in USA
Some of the needs include:
It looks something that can be interesting for this project as we can read from the IE homepage]]>
Below you can find title and abstract of the talk and bio of the speaker:
Title: Enabling Domain-Specific End-User Programming for Smart Devices
Abstract: The rapid spread of computers and other smart devices means that an increasingly large population is faced with using such devices. In many situations, some users either have to, or simply want to, make modifications to the behavior of their devices. In the extreme, some of these users want to program their devices. Considering that some of these devices are quite complex, programming them by lay persons is a challenging task. To address this challenge, we have devised a software architecture called ULD that decomposes the problem into three distinct layers: the User, the Language, and the Domain layers. The domain layer abstracts the functionality of the domain in which the device is deployed; the language layer presents a programming language specific to the domain; the user layer offers a visual environment to the end-user for programming applications for the domain. The creation of each layer requires different kinds of expertise: domain expertise, programming language expertise, and application expertise. The ULD architecture enables people with these different kinds of expertise to collaborate and combine their work. We also attempt to automate the creation of the language layer as much as possible. Our work draws on early work on end-user programming and on later work on domain-specific architectures and languages.
We believe that this work has application in a wide range of areas including smart devices, instruments, smart home devices, and Web 2.0 communities. We have implemented the architecture in a prototype demonstration with Lego Mindstorms robots.
About the speaker: Mehdi Jazayeri is professor of computer science and founding dean of the Faculty of Informatics at the University of Lugano since October 2004. Before that he was a professor and head of the Distributed Systems Group at the Technical University of Vienna (1994-2004). He worked at several startup companies in Silicon Valley before joining Hewlett-Packard Laboratories in Palo Alto for ten years (1984-94). He began his career as an assistant professor at the Computer Science Department of the University of North Carolina at Chapel Hill (1975-1980). Mehdi Jazayeri is an IEEE Fellow and was program co-chair of ICSE 2000 and program chair of ESEC-FSE 1997, the two premier international software engineering conferences.]]>