I’ve heard a lot of buzz recently around Spark. So I caught up with Ion Stoica and Mike Franklin for a call. Let me start by acknowledging some sources of confusion.
- Spark is very new. All Spark adoption is recent.
- Databricks was founded to commercialize Spark. It is very much in stealth mode …
- … except insofar as Databricks folks are going out and trying to drum up Spark adoption.
- Ion Stoica is running Databricks, but you couldn’t tell that from his UC Berkeley bio page. Edit: After I posted this, Ion’s bio was quickly updated.
- Spark creator and Databricks CTO Matei Zaharia is an MIT professor, but actually went on leave there before he ever showed up.
- Cloudera is perhaps Spark’s most visible supporter. But Cloudera’s views of Spark’s role in the world is different from the Spark team’s.
The “What is Spark?” question may soon be just as difficult as the ever-popular “What is Hadoop?” That said — and referring back to my original technical post about Spark and also to a discussion of prominent Spark user ClearStory — my try at “What is Spark?” goes something like this:
- Spark is a distributed execution engine for analytic processes …
- … which works well with Hadoop.
- Spark is distinguished by a flexible in-memory data model …
- … and farms out persistence to HDFS (Hadoop Distributed File System) or other existing data stores.
- Intended analytic use cases for Spark include:
- SQL data manipulation.
- ETL-like data manipulation.
- Streaming-like data manipulation.
- Machine learning.
- Graph analytics.
I first wrote about in-memory data management a decade ago. But I long declined to use that term — because there’s almost always a persistence story outside of RAM — and coined “memory-centric” as an alternative. Then I relented 1 1/2 years ago, and defined in-memory DBMS as
DBMS designed under the assumption that substantially all database operations will be performed in RAM (Random Access Memory)
By way of contrast:
Hybrid memory-centric DBMS is our term for a DBMS that has two modes:
- Querying and updating (or loading into) persistent storage.
These definitions, while a bit rough, seem to fit most cases. One awkward exception is Aerospike, which assumes semiconductor memory, but is happy to persist onto flash (just not spinning disk). Another is Kognitio, which is definitely lying when it claims its product was in-memory all along, but may or may not have redesigned its technology over the decades to have become more purely in-memory. (But if they have, what happened to all the previous disk-based users??)
Two other sources of confusion are:
- The broad variety of memory-centric data management approaches.
- The over-enthusiastic marketing of SAP HANA.
With all that said, here’s a little update on in-memory data management and related subjects.
- I maintain my opinion that traditional databases will eventually wind up in RAM.
- At conventional large enterprises — as opposed to for example pure internet companies — production deployments of HANA are probably comparable in number and investment to production deployments of Hadoop. (I’m sorry, but much of my supporting information for that is confidential.)
- Cloudera is emphatically backing Spark. And a key aspect of Spark is that, unlike most of Hadoop, it’s memory-centric.
- It has become common for disk-based DBMS to persist data through a “log-structured” architecture. That’s a whole lot like what you do for persistence in a fundamentally in-memory system.
- I’m also sensing increasing comfort with the strategy of committing writes as soon as they’ve been acknowledged by two or more nodes in RAM.
- I’ve never heard a story about an in-memory DBMS actually losing data. It’s surely happened, but evidently not often.
|Categories: Aerospike, Cloudera, Clustering, Databricks, Spark and BDAS, Hadoop, In-memory DBMS, Kognitio, Market share and customer counts, Memory-centric data management, SAP AG, Theory and architecture||13 Comments|
Cassandra’s reputation in many quarters is:
- World-leading in the geo-distribution feature.
- Impressively scalable.
- Hard to use.
This has led competitors to use, and get away with, sales claims along the lines of “Well, if you really need geo-distribution and can’t wait for us to catch up — which we soon will! — you should use Cassandra. But otherwise, there are better choices.”
My friends at DataStax, naturally, don’t think that’s quite fair. And so I invited them — specifically Billy Bosworth and Patrick McFadin — to educate me. Here are some highlights of that exercise.
DataStax and Cassandra have some very impressive accounts, which don’t necessarily revolve around geo-distribution. Netflix, probably the flagship Cassandra user — since Cassandra inventor Facebook adopted HBase instead — actually hasn’t been using the geo-distribution feature. Confidential accounts include:
- A petabyte or so of data at a very prominent company, geo-distributed, with 800+ nodes, in a kind of block storage use case.
- A messaging application at a very prominent company, anticipated to grow to multiple data centers and a petabyte of so of data, across 1000s of nodes.
- A 300 terabyte single-data-center telecom account (which I can’t find on DataStax’s extensive customer list).
- A huge health records deal.
- A Fortune 10 company.
DataStax and Cassandra won’t necessarily win customer-brag wars versus MongoDB, Couchbase, or even HBase, but at least they’re strongly in the competition.
DataStax claims that simplicity is now a strength. There are two main parts to that surprising assertion. Read more
|Categories: Cassandra, Clustering, Couchbase, Data models and architecture, DataStax, Facebook, HBase, Health care, Log analysis, Market share and customer counts, MongoDB and 10gen, NoSQL, Petabyte-scale data management, Specific users||10 Comments|
Relational DBMS used to be fairly straightforward product suites, which boiled down to:
- A big SQL interpreter.
- A bunch of administrative and operational tools.
- Some very optional add-ons, often including an application development tool.
Now, however, most RDBMS are sold as part of something bigger.
- Oracle has hugely thickened its stack, as part of an Innovator’s Solution strategy — hardware, middleware, applications, business intelligence, and more.
- IBM has moved aggressively to a bundled “appliance” strategy. Even before that, IBM DB2 long sold much better to committed IBM accounts than as a software-only offering.
- Microsoft SQL Server is part of a stack, starting with the Windows operating system.
- Sybase was an exception to this rule, with thin(ner) stacks for both Adaptive Server Enterprise and Sybase IQ. But Sybase is now owned by SAP, and increasingly integrated as a business with …
- … SAP HANA, which is closely associated with SAP’s applications.
- Teradata has always been a hardware/software vendor. The most successful of its analytic DBMS rivals, in some order, are:
- Netezza, a pure appliance vendor, now part of IBM.
- Greenplum, an appliance-mainly vendor for most (not all) of its existence, and in particular now as a part of EMC Pivotal.
- Vertica, more of a software-only vendor than the others, but now owned by and increasingly mainstreamed into hardware vendor HP.
- MySQL’s glory years were as part of the “LAMP” stack.
- Various thin-stack RDBMS that once were or could have been important market players … aren’t. Examples include Progress OpenEdge, IBM Informix, and the various strays adopted by Actian.
Glassbeam checked in recently, and they turn out to exemplify quite a few of the themes I’ve been writing about. For starters:
- Glassbeam has an analytic technology stack focused on poly-structured machine-generated data.
- Glassbeam partially organizes that data into event series …
- … in a schema that is modified as needed.
Glassbeam basics include:
- Founded in 2009.
- Based in Santa Clara. Back-end engineering in Bangalore.
- $6 million in angel money; no other VC.
- High single-digit customer count, …
- … plus another high single-digit number of end customers for an OEM offering a limited version of their product.
All Glassbeam customers except one are SaaS/cloud (Software as a Service), and even that one was only offered a subscription (as oppose to perpetual license) price.
So what does Glassbeam’s technology do? Glassbeam says it is focused on “machine data analytics,” specifically for the “Internet of Things”, which it distinguishes from IT logs.* Specifically, Glassbeam sells to manufacturers of complex devices — IT (most of its sales so far ), medical, automotive (aspirational to date), etc. — and helps them analyze “phone home” data, for both support/customer service and marketing kinds of use cases. As of a recent release, the Glassbeam stack can: Read more
Hortonworks did a business-oriented round of outreach, talking with at least Derrick Harris and me. Notes from my call — for which Rob Bearden didn’t bother showing up — include, in no particular order:
- Hortonworks denies advanced acquisition discussions with either Microsoft and Intel. Of course, that doesn’t exactly contradict the widespread story of Intel having made an acquisition offer. Edit: I have subsequently heard, very credibly, that the denial was untrue.
- As vendors usually do, Hortonworks denies the extreme forms of Cloudera’s suggestion that Hortonworks competitive wins relate to price slashing. But Hortonworks does believe that its license fees often wind up being lower than Cloudera’s, due especially to Hortonworks offering few extra-charge items than Cloudera.
- Hortonworks used a figure of ~75 subscription customers. Edit: That figure turns out in retrospect to have been inflated. This does not include OEM sales through, for example, Teradata, Microsoft Azure, or Rackspace. However, that does include …
- … a small number of installations hosted in the cloud — e.g. ~2 on Amazon Web Services — or otherwise remotely. Also, testing in the cloud seems to be fairly frequent, and the cloud can also be a source of data ingested into Hadoop.
- Since Hortonworks a couple of times made it seem that Rackspace was an important partner, behind only Teradata and Microsoft, I finally asked why. Answers boiled down to a Rackspace Hadoop-as-a-service offering, plus joint work to improve Hadoop-on-OpenStack.
- Other Hortonworks reseller partners seem more important in terms of helping customers consume HDP (Hortonworks Data Platform), rather than for actually doing Hortonworks’ selling for it. (This is unsurprising — channel sales rarely are a path to success for a product that is also appropriately sold by a direct force.)
- Hortonworks listed its major industry sectors as:
- Web and retailing, which it identifies as one thing.
- Health care (various subsectors).
- Financial services, which it called “competitive” in the kind of tone that usually signifies “we lose a lot more than we win, and would love to change that”.
In Hortonworks’ view, Hadoop adopters typically start with a specific use case around a new type of data, such as clickstream, sensor, server log, geolocation, or social. Read more
My clients at Aerospike are coming out with their Version 3, and as several of my clients do, have encouraged me to front-run what otherwise would be the Monday embargo.
I encourage such behavior with arguments including:
- “Nobody else is going to write in such technical detail anyway, so they won’t mind.”
- “I’ve done this before. Other writers haven’t complained.”
- “In fact, some other writers like having me go first, so that they can learn from and/or point to what I say.”
- “Hey, I don’t ask for much in the way of exclusives, but I’d be pleased if you threw me this bone.”
Aerospike 2’s value proposition, let us recall, was:
… performance, consistent performance, and uninterrupted operations …
- Aerospike’s consistent performance claims are along the lines of sub-millisecond latency, with 99.9% of responses being within 5 milliseconds, and even a node outage only borking performance for some 10s of milliseconds.
- Uninterrupted operation is a core Aerospike design goal, and the company says that to date, no Aerospike production cluster has ever gone down.
The major support for such claims is Aerospike’s success in selling to the digital advertising market, which is probably second only to high-frequency trading in its low-latency demands. For example, Aerospike’s CMO Monica Pal sent along a link to what apparently is:
- a video by a customer named Brightroll …
- … who enjoy SLAs (Service Level Agreements) such as those cited above (they actually mentioned five 9s)* …
- … at peak loads of 10-12 million requests/minute.
|Categories: Aerospike, Market share and customer counts, Memory-centric data management, NoSQL, Pricing, Web analytics||2 Comments|
Some subjects just keep coming up. And so I keep saying things like:
Most generalizations about “Big Data” are false. “Big Data” is a horrific catch-all term, with many different meanings.
Most generalizations about Hadoop are false. Reasons include:
- Hadoop is a collection of disparate things, most particularly data storage and application execution systems.
- The transition from Hadoop 1 to Hadoop 2 will be drastic.
- For key aspects of Hadoop — especially file format and execution engine — there are or will be widely varied options.
Hadoop won’t soon replace relational data warehouses, if indeed it ever does. SQL-on-Hadoop is still very immature. And you can’t replace data warehouses unless you have the power of SQL.
Note: SQL isn’t the only way to provide “the power of SQL”, but alternative approaches are just as immature.
Most generalizations about NoSQL are false. Different NoSQL products are … different. It’s not even accurate to say that all NoSQL systems lack SQL interfaces. (For example, SQL-on-Hadoop often includes SQL-on-HBase.)
I chatted yesterday with the Hortonworks gang. The main subject was Hortonworks’ approach to SQL-on-Hadoop — commonly called Stinger — but at my request we cycled through a bunch of other topics as well. Company-specific notes include:
- Hortonworks founder J. Eric “Eric14″ Baldeschwieler is no longer at Hortonworks, although I imagine he stays closely in touch. What he’s doing next is unspecified, except by the general phrase “his own thing”. (Derrick Harris has more on Eric’s departure.)
- John Kreisa still is at Hortonworks, just not as marketing VP. Think instead of partnerships and projects.
- ~250 employees.
- ~70-75 subscription customers.
Our deployment and use case discussions were a little confused, because a key part of Hortonworks’ strategy is to support and encourage the idea of combining use cases and workloads on a single cluster. But I did hear:
- 10ish nodes for a typical starting cluster.
- 100ish nodes for a typical “data lake” committed adoption.
- Teradata UDA (Unified Data Architecture)* customers sometimes (typically?) jumping straight to a data lake scenario.
- A few users in the 10s of 1000s of nodes. (Obviously Yahoo is one.)
- HBase used in >50% of installations.
- Hive probably even more than that.
- Hortonworks is seeing a fair amount of interest in Windows Hadoop deployments.
*By the way — Teradata seems serious about pushing the UDA as a core message.
Ecosystem notes, in Hortonworks’ perception, included:
- Cloudera is obviously Hortonworks’ biggest distro competitor. Next is IBM, presumably in its blue-forever installed base. MapR is barely on the radar screen; Pivotal’s likely rise hasn’t yet hit sales reports.
- Hortonworks evidently sees a lot of MicroStrategy and Tableau, and some Platfora and Datameer, the latter two at around the same level of interest.
- Accumulo is a big deal in the Federal government, and has gotten a few health care wins as well. Its success is all about security. (Note: That’s all consistent with what I hear elsewhere.)
I also asked specifically about OpenStack. Hortonworks is a member of the OpenStack project, contributes nontrivially to Swift and other subprojects, and sees Rackspace as an important partner. But despite all that, I think strong Hadoop/OpenStack integration is something for the indefinite future.
Hortonworks’ views about Hadoop 2.0 start from the premise that its goal is to support running a multitude of workloads on a single cluster. (See, for example, what I previously posted about Tez and YARN.) Timing notes for Hadoop 2.0 include:
- It’s been in preview/release candidate/commercial beta mode for weeks.
- Q3 is the goal; H2 is the emphatic goal.
- Yahoo’s been in production with YARN >8 months, and has no MapReduce 1 clusters left. (Yahoo has >35,000 Hadoop nodes.)
- The last months of delays have been mainly about sprucing up various APIs and protocols, which may need to serve for a similar multi-year period as Hadoop 1’s have. But there also was some YARN stabilization into May.
Frankly, I think Cloudera’s earlier and necessarily incremental Hadoop 2 rollout was a better choice than Hortonworks’ later big bang, even though the core-mission aspect of Hadoop 2.0 is what was least ready. HDFS (Hadoop Distributed File System) performance, NameNode failover and so on were well worth having, and it’s more than a year between Cloudera starting supporting them and when Hortonworks is offering Hadoop 2.0.
Hortonworks’ approach to doing SQL-on-Hadoop can be summarized simply as “Make Hive into as good an analytic RDBMS as possible, all in open source”. Key elements include: Read more
I’m not having a productive week, part of the reason being a hard drive crash that took out early drafts of what were to be last weekend’s blog posts. Now I’m operating from a laptop, rather than my preferred dual-monitor set-up. So please pardon me if I’m concise even by comparison to my usual standards.
- My recent posts based on surveillance news have been partly superseded by – well, by more news. Some of that news, along with some good discussion, may be found in the comment threads.
- The same goes for my recent Hadoop posts.
- The replay for my recent webinar on real-time analytics is now available. My part ran <25 minutes.
- One of my numerous clients using or considering a “real-time analytics” positioning is Sqrrl, the company behind the NoSQL DBMS Accumulo. Last month, Derrick Harris reported on a remarkable Accumulo success story – multiple US intelligence instances managing 10s of petabytes each, and supporting a variety of analytic (I think mainly query/visualization) approaches.
- Several sources have told me that MemSQL’s Zynga sale is (in part) for Membase replacement. This is noteworthy because Zynga was the original pay-for-some-of-the-development Membase customer.
- More generally, the buzz out of Couchbase is distressing. Ex-employees berate the place; job-seekers check around and then decide not to go there; rivals tell me of resumes coming out in droves. Yes, there’s always some of that, even at obviously prospering companies, but this feels like more than the inevitable low-level buzz one hears anywhere.
- I think the predictive modeling state of the art has become:
- Cluster in some way.
- Model separately on each cluster.
- And if you still want to do something that looks like a regression – linear or otherwise – then you might want to use a tool that lets you shovel training data in WITHOUT a whole lot of preparation* and receive a model back out. Even if you don’t accept that as your final model, it can at least be a great guide to feature selection (in the statistical sense of the phrase) and the like.
- Champion/challenger model testing is also a good idea, at least if you’re in some kind of personalization/recommendation space, and have enough traffic to test like that.**
- Most companies have significant turnover after being acquired, perhaps after a “golden handcuff” period. Vertica is no longer an exception.
- Speaking of my clients at HP Vertica – they’ve done a questionable job of communicating that they’re willing to price their product quite reasonably. (But at least they allowed me to write about $2K/terabyte for hardware/software combined.)
- I’m hearing a little more Amazon Redshift buzz than I expected to. Just a little.
- StreamBase was bought by TIBCO. The rumor says $40 million.
*Basic and unavoidable ETL (Extract/Transform/Load) of course excepted.
**I could call that ABC (Always Be Comparing) or ABT (Always Be Testing), but they each sound like – well, like The Glove and the Lions.