Introduction to Spark, Shark, BDAS and AMPLab
UC Berkeley’s AMPLab is working on a software stack that:
- Is meant (among other goals) to improve upon Hadoop …
- … but also to interoperate with it, and which in fact …
- … uses significant parts of Hadoop.
- Seems to have the overall name BDAS (Berkeley Data Analytics System).
The whole thing has $30 million in projected funding (half government, half industry) and a 6-year plan (which they’re 2 years into).
Specific projects of note in all that include:
- Mesos, a cluster manager. I don’t know much about Mesos, but it seems to be in production use, most notably at Twitter supporting Storm.
- Spark, a replacement for MapReduce and the associated execution stack.
- Shark, a replacement for Hive.
| Categories: ClearStory Data, Databricks, Spark and BDAS, Hadoop, MapReduce, Parallelization, Specific users, SQL/Hadoop integration | 11 Comments |
Some trends that will continue in 2013
I’m usually annoyed by lists of year-end predictions. Still, a reporter asked me for some, and I found one kind I was comfortable making.
Trends that I think will continue in 2013 include:
Growing attention to machine-generated data. Human-generated data grows at the rate business activity does, plus 0-25%. Machine-generated data grows at the rate of Moore’s Law, also plus 0-25%, which is a much higher total. In particular, the use of remote machine-generated data is becoming increasingly real.
Hadoop adoption. Everybody has the big bit bucket use case, largely because of machine-generated data. Even today’s technology is plenty good enough for that purpose, and hence justifies initial Hadoop adoption. Development of further Hadoop technology, which I post about frequently, is rapid. And so the Hadoop trend is very real.
Application SaaS. The on-premises application software industry has hopeless problems with product complexity and rigidity. Any suite new enough to cut the Gordian Knot is or will be SaaS (Software as a Service).
Newer BI interfaces. Advanced visualization — e.g. Tableau or QlikView — and mobile BI are both hot. So, more speculatively, are “social” BI (Business Intelligence) interfaces.
Price discounts. If you buy software at 50% of list price, you’re probably doing it wrong. Even 25% can be too high.
MySQL alternatives. NoSQL and NewSQL products often are developed as MySQL alternatives. Oracle has actually done a good job on MySQL technology, but now its business practices are scaring companies away from MySQL commitments, and newer short-request SQL DBMS are ready for use.
| Categories: Business intelligence, Hadoop, MySQL, NewSQL, NoSQL, Open source, Oracle, Pricing, Software as a Service (SaaS), Surveillance and privacy | 3 Comments |
Amazon Redshift and its implications
Merv Adrian and Doug Henschen both reported more details about Amazon Redshift than I intend to; see also the comments on Doug’s article. I did talk with Rick Glick of ParAccel a bit about the project, and he noted:
- Amazon Redshift is missing parts of ParAccel, notably the extensibility framework.
- ParAccel did some engineering to make its DBMS run better in the cloud.
- Amazon did some engineering in the areas it knows better than ParAccel — cloud provisioning, cloud billing, and so on.
“We didn’t want to do the deal on those terms” comments from other companies suggest ParAccel’s main financial take from the deal is an already-reported venture investment.
The cloud-related engineering was mainly around communications, e.g. strengthening error detection/correction to make up for the lack of dedicated switches. In general, Rick seemed more positive on running in the (Amazon) cloud than analytic RDBMS vendors have been in the past.
So who should and will use Amazon Redshift? For starters, I’d say: Read more
ParAccel update
In connection with Amazon’s Redshift announcement, ParAccel reached out, and so I talked with them for the first time in a long while. At the highest level:
- ParAccel now has 60+ customers, up from 30+ two years ago and 40ish soon thereafter.
- ParAccel is now focusing its development and marketing on analytic platform capabilities more than raw database performance.
- ParAccel is focusing on working alongside other analytic data stores — relational or Hadoop — rather than supplanting them.
There wasn’t time for a lot of technical detail, but I gather that the bit about working alongside other data stores:
- Is relatively new.
- Works via SELECT statements that reach out to the other data stores.
- Is called “on-demand integration”.
- Is built in ParAccel’s extensibility/analytic platform framework.
- Uses HCatalog when reaching into Hadoop.
Also, it seems that ParAccel:
- Is in the early stages of writing its own analytic functions.
- Bundles Fuzzy Logix and actually has some users for that.
| Categories: Amazon and its cloud, Cloud computing, Data warehousing, Hadoop, Market share and customer counts, ParAccel, Predictive modeling and advanced analytics, Specific users | 5 Comments |
Are column stores really better at compression?
A consensus has evolved that:
- Columnar compression (i.e., value-based compression) compresses better than block-level compression (i.e., compression of bit strings).
- Columnar compression can be done pretty well in row stores.
Still somewhat controversial is the claim that:
- Columnar compression can be done even better in column stores than in row-based systems.
A strong plausibility argument for the latter point is that new in-memory analytic data stores tend to be columnar — think HANA or Platfora; compression is commonly cited as a big reason for the choice. (Another reason is that I/O bandwidth matters even when the I/O is from RAM, and there are further reasons yet.)
One group that made the in-memory columnar choice is the Spark/Shark guys at UC Berkeley’s AMP Lab. So when I talked with them Thursday (more on that another time, but it sounds like cool stuff), I took some time to ask why columnar stores are better at compression. In essence, they gave two reasons — simplicity, and speed of decompression.
In each case, the main supporting argument seemed to be that finding the values in a column is easier when they’re all together in a column store. Read more
| Categories: Columnar database management, Database compression, Databricks, Spark and BDAS, In-memory DBMS, Netezza | 10 Comments |
Notes on Microsoft SQL Server
I’ve been known to gripe that covering big companies such as Microsoft is hard. Still, Doug Leland of Microsoft’s SQL Server team checked in for phone calls in August and again today, and I think I got enough to be worth writing about, albeit at a survey level only,
Subjects I’ll mention include:
- Hadoop
- Parallel Data Warehouse
- PolyBase
- Columnar data management
- In-memory data management (Hekaton)
One topic I can’t yet comment about is MOLAP/ROLAP, which is a pity; if anybody can refute my claim that ROLAP trumps MOLAP, it’s either Microsoft or Oracle.
Microsoft’s slides mentioned Yahoo refining a 6 petabyte Hadoop cluster into a 24 terabyte SQL Server “cube”, which was surprising in light of Yahoo’s history as an Oracle reference.
Couchbase 2.0
My clients at Couchbase checked in.
- After multiple delays, Couchbase 2.0 is well into beta, with general availability being delayed by the holiday season as much as anything else.
- Couchbase (the company) now has >350 subscription customers, almost all for Couchbase (the product) — which is to say for what was known as Membase, which is basically a persistent version of Memcached.
- There also are many users of open source Couchbase, most famously LinkedIn.
- Orbitz is a much-mentioned flagship paying Couchbase customer.
- Couchbase customers mainly seem to be replacing a caching layer, Memcached or otherwise.
- Couchbase headcount is just under 100.
The big changes in Couchbase 2.0 versus the previous (1.8.x) version are:
- JSON storage, including secondary indexes.
- Multi-data-center replication.
- A back-end change from SQLite to a heavily forked version of CouchDB, called Couchstore.
Couchbase 2.0 is upwards-compatible with prior versions of Couchbase (and hence with Memcached), but not with CouchDB.
Technology notes on Couchbase 2.0 include: Read more
| Categories: Basho and Riak, Cache, Cassandra, Clustering, Couchbase, MapReduce, Market share and customer counts, MongoDB, NoSQL, Open source, Structured documents | 5 Comments |
Incremental MapReduce
My clients at Cloudant, Couchbase, and 10gen/MongoDB (Edit: See Alex Popescu’s comment below) all boast the feature incremental MapReduce. (And they’re not the only ones.) So I feel like making a quick post about it. For starters, I’ll quote myself about Cloudant:
The essence of Cloudant’s incremental MapReduce seems to be that data is selected only if it’s been updated since the last run. Obviously, this only works for MapReduce algorithms whose eventual output can be run on different subsets of the target data set, then aggregated in a simple way.
These implementations of incremental MapReduce are hacked together by teams vastly smaller than those working on Hadoop, and surely fall short of Hadoop in many areas such as performance, fault-tolerance, and language support. That’s a given. Still, if the jobs are short and simple, those deficiencies may be tolerable.
A StackOverflow thread about MongoDB’s version of incremental MapReduce highlights some of the implementation challenges.
But all practicality aside, let’s return to the point that incremental MapReduce only works for some kinds of MapReduce-based algorithms, and consider how much of a limitation that really is. Looking at the Map steps sheds a little light: Read more
| Categories: Cloudant, Couchbase, EAI, EII, ETL, ELT, ETLT, Hadoop, MapReduce, MongoDB, RDF and graphs | 1 Comment |
Analytic application subsystems
Imagine a website whose purpose is to encourage consumers to take actions — for example to click on an ad, click on the next page, or actually make a purchase. Best practices for such a site include:
- An ever-evolving user experience, informed by — among other factors — creativity, brand identity, the vendor’s evolving product line itself, and …
- … predictive modeling.
- Personalization based on predictive modeling.
Those predictive models themselves will keep changing, because:
- Organizations learn.
- Consumer tastes change.
- More or different kinds of data keep becoming available.
In that situation, what would it mean to offer the website owner a predictive modeling “application”? Read more
Do you need an analytic RDBMS?
I can think of seven major reasons not to use an analytic RDBMS. One is good; but the other six seem pretty questionable, niche circumstances excepted, especially at this time.
The good reason to not have an analytic RDBMS is that most organizations can run perfectly well on some combination of:
- SaaS (Software as a Service).
- A low-volume static website.
- A network focused on office software.
- A single cheap server, likely running a single instance of a general-purpose RDBMS.
Those enterprises, however, are generally not who I write for or about.
The six bad reasons to not have an analytic RDBMS all take the form “Can’t some other technology do the job better?”, namely:
- A data warehouse that’s just another instance of your OLTP (OnLine Transaction Processing) RDBMS. If your problem is that big, it’s likely that a specialized analytic RDBMS will be more cost-effective and generally easier to deal with.
- MOLAP (Multi-Dimensional OnLine Analytic Processing). That ship has sailed … and foundered … and been towed to drydock.
- In-memory BI. QlikView, SAP HANA, Oracle Exalytics, and Platfora are just four examples of many. But few enterprises will want to confine their analytics to such data as fits affordably in RAM.
- Non-tabular* approaches to investigative analytics. There are many examples in the Hadoop world — including the recent wave of SQL add-ons to Hadoop — and some in the graph area as well. But those choices will rarely suffice for the whole job, as most enterprises will want better analytic SQL performance for (big) parts of their workloads.
- Tighter integration of analytics and OLTP (OnLine Transaction Processing). Workday worklets illustrate that business intelligence/OLTP integration is a really good idea. And it’s an idea that Oracle and SAP can be expected to push heavily, when they finally get their product acts together. But again, that’s hardly all the analytics you’re going to want to do.
- Tighter integration of analytics and other short-request processing. An example would be maintaining a casual game’s leaderboard via a NoSQL write-optimized database. Yet again, that’s hardly all the analytics a typical enterprise will want to do.