In our latest release of SMB, we spent a fair amount of time testing and minimizing windows where data corruption can occur in our associative memory. The testing involved killing various processes during periods of activity. But rather than a hit-or-miss approach of randomly killing SMB processes and seeing if any sort of recovery was needed, we decided to build a JVMKiller that gave us control of when and how the JVM goes down. We are now able to target specific “high risk” (write operations) areas. Like everything else in our system, the JVMKiller is configured as a JavaBean. The bean has the ability to execute any number of possible kill events, as well as, targeting both specific logical and physical components of the system, given a condition. For example, we could target a certain event, like the start of a phase-2 save event:
(more…)
Posts Tagged ‘development’
JVMKiller
Tuesday, June 22nd, 2010 by Jim Fleming
Tags: development, smb, testing
Posted in SaffronMemoryBase | No Comments »
SaffronSierra & Gmail Classification
Friday, April 30th, 2010 by Jared Peterson
I’ve recently added some sample code to our “examples” repository that demonstrates how to use SaffronMemoryBase running on SaffronSierra to do basic email classification. The example leverages the convenience of labels within Gmail to provide the “labels” for classifying future emails.
If you have a Gmail account (or a Google Apps account) then you already know that as emails come in you can associate them with labels. You might have labels such as “accounts”, “soccer”, “music”, “work”, etc… (those are some of mine anyway). As I started thinking about building an email classification example it occurred to me that the labels within Gmail would provide a nice, easy way of doing classification. By “labeling” emails in Gmail I’ve already made the statement, “This email is about work”, or “This email is about soccer”. Why not leverage that hard work?
Tags: classifications, development, email, gmail, Java, SaffronSierra, smb
Posted in SaffronMemoryBase, SaffronSierra | No Comments »
SaffronSierra Refresh
Friday, April 16th, 2010 by Jared Peterson
We just pushed out a minor refresh to SaffronSierra. As a user you won’t notice much that’s different (hopefully). This refresh brings with it a couple of changes. One that users will see, and another they won’t. Let me talk about both briefly:
- Nightly Running Reminders – SaffronSierra is a pay-as-you-go service. You pay for every hour your service is running. SaffronSierra makes it really simple (you click a button) to start a SaffronMemoryBase system. It’s so simple in fact that it’s easy to start it and forget that it’s running. Now SaffronSierra will send email reminders every night to users who’s SaffronSierra service is running. If you actually want your service to be running 24/7 (which is totally fine by the way), and you don’t want to receive these reminders you can login and update your user profile.
- New Amazon AWS Java Libraries – SaffronSierra uses Amazon AWS as it’s backend infrastructure for running services. The java library we originally used to integrate with AWS has been deprecated. So, we ported SaffronSierra over to the new java library. This actually was a pretty painless process. There are some differences between the two libraries, but as I told someone recently “it’s different in the ways that it needs to be different.” Hopefully you won’t notice this change and things just keep humming along.
We’ll keep the updates coming, stay tuned.
Tags: amazon, development, ec2, Java, SaffronSierra
Posted in SaffronMemoryBase, SaffronSierra | No Comments »
TweetDive Source Code
Wednesday, April 14th, 2010 by Jared Peterson
I’m happy to announce that we’ve made the TweetDive source code publicly available (finally). I know a number of people have been waiting to get their hands on it. For details on how to grab it from Subversion take a look at this page.
We hope this code will provide a quick-start for developers getting started with SaffronSierra. TweetDive uses a number of different REST APIs available on SaffronSierra. You can use it as a reference as you build your own stuff, or you can extend TweetDive with new features. If you’d like to see your changes incorporated into the running version of TweetDive please contact us. We’d be happy to incorporate patches.
Tags: applications, development, REST, SaffronSierra, TweetDive, twitter
Posted in SaffronMemoryBase, SaffronSierra, TweetDive | No Comments »
Partnering with Saffron
Monday, April 12th, 2010 by Gayle Sheppard
We’re hearing it every day: Customers want better, more advanced analytic capabilities.
Customers are telling us they want to unify and apply advanced data analytics to disparate data sources –– semantic web, enterprise data, sensor data, structured and unstructured data, for example –– to improve business decisions and results.
It’s music to our ears because that’s what our associative memory technology does: We provide “beyond BI” capabilities, in a single tool, for sense-making and decision support. We can help our customers make better sense of things and make better decisions because we expose all the relevant experiences –– connections, similarities, prior outcomes for example –– across all the data in real time.
Tags: applications, development, Partners
Posted in Natural Intelligence, Partners, SaffronSierra | No Comments »
What is SaffronSierra (in plain english)?
Wednesday, February 17th, 2010 by Jared Peterson
I recently received a request to describe SaffronSierra “in plain english”, this post will attempt to do just that.
SaffronSierra is a cloud-based, hosted version of SaffronMemoryBase. SaffronSierra enables developers to easily leverage the analytical powers of SaffronMemoryBase using a simple REST API. Using SaffronSierra, you don’t have to worry about downloading, installing and configuring software. You simply login to your SaffronSierra console and press “Start”. Within minutes, you will have a private instance of SaffronSierra ready for your use. Once your service is available you can start putting your data into SaffronSierra and powering your applications with advanced analytics. We hope that SaffronSierra enables people to quickly realize the benefits of SaffronMemoryBase in a low-cost, low-risk way.
Tags: amazon, development, REST, SaffronSierra, TweetDive, twitter
Posted in SaffronSierra | No Comments »
Observing & Querying Tweets (Part 3)
Wednesday, January 13th, 2010 by Jared Peterson
In my previous posts I’ve discussed how to fetch data from Twitter and massage it into the form that is needed when observing data in Sierra. I also talked about how to “POST” that data to Sierra so that it’s available to query. In this post I’m going to discuss the process of actually querying that data once it’s in Sierra.
Before you read on you may want to browse over the Sierra documentation. These docs list all of the REST calls that are available on Sierra. In my example below I’ll be using the “connections” call.
I should also note that I’ve made some additions to the source code that I’ve used across all of these posts. That code is available along with the rest of the Sierra sample code. In particular I’ve made the code aware of command-line options that specify whether you are “observing” (-o) or “querying” (-q). You can also define all of the needed usernames and passwords via the command-line as well.
Tags: development, Groovy, json, REST, SaffronSierra, twitter
Posted in SaffronSierra | 1 Comment »
Ingesting Data into Sierra
Monday, January 11th, 2010 by Chet Patel
After signing up for a Sierra account, a few simple steps can be taken to push your data into Sierra and then analyze your data.
First, you’ll need to grab our sample code. You can find the instructions for downloading our sample code here.
Once you have downloaded the sample code, look for the Ingest.java class (in the src/java directory). This is a command line tool that can be run from your JAVA IDE or as a compiled java application.
Tags: development, REST, SaffronSierra
Posted in SaffronSierra | No Comments »
Observing & Querying Tweets (Part 2)
Monday, December 28th, 2009 by Jared Peterson
In my last post I walked through grabbing tweets from Twitter using their API. We then took those tweets and built the necessary XML “resource” for pushing that data into Saffron Sierra. In this post I’m going to cover the process of actually getting the data into (observing) Sierra.
To start, I should probably cover a minor modification I made to the code from the last post as I worked through this post. Last time, I created one XML resource for Sierra that contained all of the tweets from my timeline. After some testing, and thinking, I decided to put each tweet into it’s own separate XML resource. I think this is cleaner, and it’s the “supported” path for Sierra anyway.
Tags: development, Groovy, REST, SaffronSierra, twitter
Posted in SaffronSierra | 2 Comments »
Observing & Querying Tweets (Part 1)
Monday, November 9th, 2009 by Jared Peterson
As we look to build compelling examples using Saffron Sierra we’ve often talked about using Twitter as a datasource. If we could get Twitter data/tweets into Sierra then we could builds lots on interesting analytical capability using Sierra’s APIs. To enable our exploration, and others, I’ve decided to start a little series (probably 3 parts) of blog post about observing and querying Twitter data in Sierra.
In this first part I’m going to cover the process of grabbing a Twitter feed and building the necessary resource XML for Sierra observations. For now, you can think of a “resource” in Sierra speak as our method of “inserting” data into the system. For my example I’m going to be using Groovy . This code should be easily replicated in your language of choice. I choose Groovy because I’m familiar with it and it provides a few tools that make this kind of stuff really easy.
Tags: development, Groovy, REST, SaffronSierra, twitter
Posted in SaffronSierra | 5 Comments »
