Good thing our US government is not at all influenced by religious fanatics who would, let's say, deny things like evolution or global warming. Oh wait...
Java is most certainly not dead, but it has transformed. The language itself is not as much relevant as the JVM abstraction, since now you can code in Python, Clojure, Scala, Ruby, to bytecode. This minimum common denominator becomes immensely important when deploying applications to the cloud, since you can find tons of providers that can run applications on JVM, and you also get for (almost) free a ton of tools to monitor your apps with JMX (debugging, profiling, etc). Hadoop and its ecosystem, Spark, Kafka, Storm and a lot of other cloud oriented software are written on the JVM. Nobody wants to care on which hardware you're running for most applications (that is, if you don't need a GPU, but that's being commoditized too).
I agree that the problem is that most companies don't know how to run it and it's left to bigger organizations that 1) have the expertise in house and 2) actually need the added complexity. Understanding which pieces of the ecosystem you need, how to deploy and running them in a production environment can be daunting, not to mention all the different possibilities of which cloud provider to use, which services, etc.
Cloudera and Hortonworks are capitalizing on it basically helping sorting out this complexity with consultants, and training, but since this business model scales with the number of employees, they are not scaling up that fast, also because there are not enough skilled engineers in the field. I personally interviewed several self proclaimed 'hadoop engineers' who had worked on hadoop for a year or more and yet didn't know what happens in the shuffle phase.
Another distinction to make is that Hadoop has now three major components: HDFS, YARN and map/reduce. Maybe Map/Reduce is losing its relevance as a hadoop component, as Tez/Shark/Flink advance, but should be noted that under the hood they use basically the same abstraction on parallelization, they just make better use of resources (especially memory), but they are not replacing HDFS not YARN. Mesos could be used in alternative of YARN, but I don't see any competitor for HDFS yet.
So, I would not say that Hadoop is being replaced, but more extended and to use a botanical analogy, beside growing, it's also being grafted on (flink,spark, cassandra, etc...).
The Prime Minister of Italy owns the largest Italian publishing house
He doesn't just own a lot of the media but also the biggest advertising company, Mediaset. In fact, there was also a similar lawsuit against youtube.
The prime minister and his lawyer posse has been quite active. They just sued a rival newspaper for asking questions that he deemed libelous: (in italian) http://www.repubblica.it/2009/08/sezioni/politica/berlusconi-divorzio-22/causa-domande/causa-domande.html
I can't help thinking that a former austronat may be just trying to help NASA to get more funding. Speaking about aliens may just bring enough media attention and get people worried enough to get space missions back on the politician radar.
Don't forget that NASA funding has been slashed dramatically and for a couple of years after the shuttle is retired we'll need russian rockets to reach the ISS, until the ARES rockets are ready.
I agree that the article misses many points. Comparing MapReduce to DBMS it's like comparing apples to oranges. In the article it says MapReduce may be a good idea for writing certain types of general-purpose computations. Thing is, MapReduce works where traditional databases don't. DBMS are general purpose and deal with concurrency, and have a Read/Insert/Update/Delete paradigm: at any time, you could perform any of these operations on any record. MapReduce is specialized on performing calculations (i.e. reading) on a huge amount of data on a cluster, with performance that increases almost linearly with the number of nodes. Also, the article cites MapReduce only, but it doesn't mention that MapReduce works hand-in-hand with a DFS (Distributed File System). Try to load 1 Terabyte of data into Oracle (or any other DBMS) daily and let me go how it goes....where are you going to store it? If your data is read once, read many (which is the case when you're dealing with logs, like search event logs) you can make assumptions that simplify your storage and processing a big deal (indexes are easier to maintain, etc). That leads to a inexpensive, commodity hardware cluster distributed file system, which is the real key here, as cheap, scalable processing power and storage space are both paramount for growth. It's kind of useless to talk about MapReduce without talking about DFS.
Actually, you're right. I am an italian living in the US and I am following this matter closely. Basically the first draft of the law that was presented to the parlament was written in an unclear way that could have been interpreted as in this article: that EVERY blog or site needed to be registered. After the complaints from the bloggers, the government already said they would change the text, as their intent was merely extend to the digital world the laws that exist for newspapers, so AFAIK just sites who provide news and information commercially and with a certain peridiocity will need to be registered.
While in the past the driving force inside the technology was the language (cobol, fortran, C, C++) now we're moving towards platforms and frameworks as a whole. The two contenders are, of course,.NET and Java with its J2EE, JME etc. As complexity of information system grows, I see PERL more and more relegated to the role of 'glue' and 'toolbox' , with the main infrastructure developed in one of those two technologies. This will happen for big sites, smaller sites will still benefit from PERL. I also see PERL as a great testing scripting language. This is my vision of PERL's future...what's yours ?
I think there's a mistake. In the specs in Sony's website they say the resolution is 320x480. That's half VGA. I am looking forward to try one of these toys, they look so sweet...
Slashdot Mirror
Good thing our US government is not at all influenced by religious fanatics who would, let's say, deny things like evolution or global warming. Oh wait...
Java is most certainly not dead, but it has transformed. The language itself is not as much relevant as the JVM abstraction, since now you can code in Python, Clojure, Scala, Ruby, to bytecode. This minimum common denominator becomes immensely important when deploying applications to the cloud, since you can find tons of providers that can run applications on JVM, and you also get for (almost) free a ton of tools to monitor your apps with JMX (debugging, profiling, etc).
Hadoop and its ecosystem, Spark, Kafka, Storm and a lot of other cloud oriented software are written on the JVM. Nobody wants to care on which hardware you're running for most applications (that is, if you don't need a GPU, but that's being commoditized too).
I agree that the problem is that most companies don't know how to run it and it's left to bigger organizations that 1) have the expertise in house and 2) actually need the added complexity.
Understanding which pieces of the ecosystem you need, how to deploy and running them in a production environment can be daunting, not to mention all the different possibilities of which cloud provider to use, which services, etc.
Cloudera and Hortonworks are capitalizing on it basically helping sorting out this complexity with consultants, and training, but since this business model scales with the number of employees, they are not scaling up that fast, also because there are not enough skilled engineers in the field. I personally interviewed several self proclaimed 'hadoop engineers' who had worked on hadoop for a year or more and yet didn't know what happens in the shuffle phase.
Another distinction to make is that Hadoop has now three major components: HDFS, YARN and map/reduce. Maybe Map/Reduce is losing its relevance as a hadoop component, as Tez/Shark/Flink advance, but should be noted that under the hood they use basically the same abstraction on parallelization, they just make better use of resources (especially memory), but they are not replacing HDFS not YARN. Mesos could be used in alternative of YARN, but I don't see any competitor for HDFS yet.
So, I would not say that Hadoop is being replaced, but more extended and to use a botanical analogy, beside growing, it's also being grafted on (flink,spark, cassandra, etc...).
...article doesn't mention ethnicity, but from the treatment, it sounds like the kid is black...
The Prime Minister of Italy owns the largest Italian publishing house
He doesn't just own a lot of the media but also the biggest advertising company, Mediaset. In fact, there was also a similar lawsuit against youtube.
The prime minister and his lawyer posse has been quite active. They just sued a rival newspaper for asking questions that he deemed libelous: (in italian) http://www.repubblica.it/2009/08/sezioni/politica/berlusconi-divorzio-22/causa-domande/causa-domande.html
I can't help thinking that a former austronat may be just trying to help NASA to get more funding.
Speaking about aliens may just bring enough media attention and get people worried enough to get space missions back on the politician radar.
Don't forget that NASA funding has been slashed dramatically and for a couple of years after the shuttle is retired we'll need russian rockets to reach the ISS, until the ARES rockets are ready.
I agree that the article misses many points.
Comparing MapReduce to DBMS it's like comparing apples to oranges.
In the article it says MapReduce may be a good idea for writing certain types of general-purpose computations. Thing is, MapReduce works where traditional databases don't.
DBMS are general purpose and deal with concurrency, and have a Read/Insert/Update/Delete paradigm: at any time, you could perform any of these operations on any record.
MapReduce is specialized on performing calculations (i.e. reading) on a huge amount of data on a cluster, with performance that increases almost linearly with the number of nodes.
Also, the article cites MapReduce only, but it doesn't mention that MapReduce works hand-in-hand with a DFS (Distributed File System). Try to load 1 Terabyte of data into Oracle (or any other DBMS) daily and let me go how it goes....where are you going to store it?
If your data is read once, read many (which is the case when you're dealing with logs, like search event logs) you can make assumptions that simplify your storage and processing a big deal (indexes are easier to maintain, etc). That leads to a inexpensive, commodity hardware cluster distributed file system, which is the real key here, as cheap, scalable processing power and storage space are both paramount for growth. It's kind of useless to talk about MapReduce without talking about DFS.
Actually, you're right.
I am an italian living in the US and I am following this matter closely.
Basically the first draft of the law that was presented to the parlament was written in an unclear way that could have been interpreted as in this article: that EVERY blog or site needed to be registered.
After the complaints from the bloggers, the government already said they would change the text, as their intent was merely extend to the digital world the laws that exist for newspapers, so AFAIK just sites who provide news and information commercially and with a certain peridiocity will need to be registered.
While in the past the driving force inside the technology was the language (cobol, fortran, C, C++) now we're moving towards platforms and frameworks as a whole. The two contenders are, of course, .NET and Java with its J2EE, JME etc.
As complexity of information system grows, I see
PERL more and more relegated to the role of 'glue' and 'toolbox' , with the main infrastructure developed in one of those two technologies. This will happen for big sites, smaller sites will still benefit from PERL.
I also see PERL as a great testing
scripting language. This is my vision of PERL's future...what's yours ?
I think there's a mistake.
In the specs in Sony's website
they say the resolution is 320x480.
That's half VGA. I am looking forward to try one of these toys, they look so sweet...
Yeah, it's Genoa, not Milan (or Genova in italian).
That's where the G8 took place. In Italy was quite a big deal.