|By Michael Kopp||
|April 29, 2013 10:15 AM EDT||
Over the last couple of months I have been talking to more and more customers who are either bringing their Hadoop clusters into production or have already done so and are now getting serious about operations. This leads to some interesting discussions about how to monitor Hadoop properly and one thing pops up quite often: Do they need anything beyond Ganglia? If yes, what should they do beyond it?
As in every other system, monitoring in a Hadoop environment starts with the basics: System Metrics - CPU, Disk, Memory you know the drill. Of special importance in a Hadoop system is a well-balanced cluster; you don't want to have some nodes being much more (or less) utilized then others. Besides CPU and memory utilization, Disk utilization and of course I/O throughput is of high importance. After all the most likely bottleneck in a Big Data system is I/O - either with ingress (network and disk), moving data around (e.g., MapReduce shuffle on the network) and straightforward read/write to disk.
The problem in a Hadoop system is of course its size. Nothing new for us, some of our customers monitor well beyond 1000+ JVMs with CompuwareAPM. The "advantage" in a Hadoop system is its relative conformity - every node looks pretty much like the other. This is what Ganglia leverages.
Cluster Monitoring with Ganglia
What Ganglia is very good at is providing an overview over how a cluster is utilized. The load chart is particularly interesting:
This chart shows the CPU load on a 1000 Server cluster that has roughly 15.000 CPUs
It tells us the number of available cores in the system and the number of running processes (in theory a core can never handle more than one process at a time) and the 1-min load average. If the system is getting fully utilized the 1-min load average would approach the total number of CPUs. Another view on this is the well-known CPU utilization chart:
CPU Utilization over the last day. While the utilization stays well below 10% we see a lot of I/O wait spikes.
While the load chart gives a good overall impression of usage, the utilization tells us the story of how the CPUs are used. While typical CPU charts show a single server, Ganglia specializes in showing whole clusters (the picture shows CPU usage of a 1000 machine cluster). In the case of the depicted chart we see that the CPUs are experiencing a lot of I/O wait spikes, which points toward heavy disk I/O. Basically it seems the disk I/O is the reason that we cannot utilize our CPU better at these times. But in general our cluster is well underutilized in terms of CPU.
Trends are also easy to understand, as can be seen in this memory chart over a year.
Memory capacity and usage over a year
All this looks pretty good, so what is missing? The "so what" and "why" is what is missing. If my memory demand is growing, I have no way of knowing why it is growing. If the CPU chart tells me that I spend a lot of time waiting, it does not tell what to do, or why that is so? These questions are beyond the scope of Ganglia.
What about Hadoop specifics?
Ganglia also has a Hadoop plugin, which basically gives you access to all the usual Hadoop metrics (unfortunately a comprehensive list of Hadoop metrics is really hard to find, appreciate if somebody commented the link). There is a good explanation on what is interesting on Edward Caproli's page: JoinTheGrid. Basically you can use those metrics to monitor the capacity and usage trends of HDFS and the NameNodes and also how many jobs, mappers and reducers are running.
Capacity of the DataNodes over time
Capacity of the Name Nodes over time
The DataNode Operations give me an impression of I/O pressure on the Hadoop cluster
All these charts can of course be easily built in any modern monitoring or APM solution like CompuwareAPM, but Ganglia gives you a simple starting point; and it's Free as in Beer.
What's missing again is the so what? If my jobs are running a lot longer than yesterday, what should I do? Why do they run longer? A Hadoop expert might dig into 10 different charts around I/O and Network, spilling, look at log files among other things and try an educated guess as to what might be the problem. But we aren't all experts, neither do we have the time to dig into all of these metrics and log files all the time.
This is the reason that we and our customers are moving beyond Ganglia - to solve the "Why" and "So What" within time constraints.
Beyond the Basics #1 - Understanding Cluster Utilization
A use case that we get from customers is that they want to know which users or which pools (in case of the fair scheduler) are responsible for how much of the cluster utilization. LinkedIn just released White Elephant, a tool that parses MapReduce logs and builds some nice dashboards and shows you which of your users occupy how much of your cluster. This is of course based on log file analysis and thus okay for analysis but not for monitoring. With proper tools in place we can do the same thing in near real time.
The CPU Usage in the Hadoop Cluster on per User basis
In this example I wanted to monitor which user consumed how much of my Amazon EMR cluster. If we see a user or pool that occupies a lot of the cluster we can course also see which jobs are running and how much of the cluster they occupy.
The CPU Usage in the Hadoop Cluster on per Job basis
And this will also tell us if that job has always been there, and just uses a lot more resources now. This would be our cue to start analyzing what has changed.
Beyond the Basics #2 - Understanding why my jobs are slow(er)
If we want to understand why a job is slow we need to look at a high-level break down first.
In which phase of the map reduce do we spend the most time, or did we spend more time than yesterday? Understanding these timings in context with the respective job counters, like Map Input or Spilled Records, helps us understand why the phase took longer.
Overview of the time spent in different phases and the respective input/output counters
At this point we will already have a pretty good idea as to what happened. We either simply have more data to crunch (more input data) or a portion of the MapReduce job consumes more CPU (code change?) or we spill more records to disk (code change or Hadoop config change?). We might also detect an unbalanced cluster in the performance breakdown.
This job is executing nearly exclusively on a single node instead of distributing
In this case we want to check whether all the involved nodes processed the same amount of data
Here we see that there is a wide range from minimum, average to maximum on mapped input and output records. The data is not balanced
or if the difference can again be found in the code (different kinds of computations). If we are running against HBase we might of course have an issue with HBase performance or distribution.
At the beginning of the job only a single HBase region Server consumes CPU while all others remain idle
On the other hand, if a lot of mapping time is spent in the garbage collector then you should maybe invest in larger JVMs.
The Performance Breakdown of this particular job shows considerable time in GC suspension
If spilling data to disk is where we spend our time, we should take a closer look at that phase. It might turn out that all of our time is spent on disk wait.
If the Disk were the bottleneck we would see it on disk I/O here
Now if disk write is our bottleneck, then really the only thing that we can do is reduce the map output records. Adding a combiner will not reduce the disk write (it will actually increase it, read here). In other words combining only optimizes the shuffle phase, thus the amount of data sent over the network, but not spill time!!
And at the very detailed level we can look at single task executions and understand in detail what is really going on.
The detailed data about each Map, Reduce Task Atttempt as well as the spills and shuffles
Ganglia is a great tool for high-level monitoring of your Hadoop cluster utilization, but it is not enough. The fact that everybody is working on additional means to understand the Hadoop cluster (Hortonworks with Ambari, Cloudera with their Manager, LinkedIn with White Elephant, the Star Fish project...) shows that there is a lot more needed beyond simple monitoring. Even those more advanced monitoring tools are not always answering the "why" though, which is what we really need to do. This is where the Performance Management discipline can add a lot of value and really help you get the best out of your Hadoop cluster. In other words don't just run Hadoop jobs at scale, run them efficiently and at scale!
In the next five to ten years, millions, if not billions of things will become smarter. This smartness goes beyond connected things in our homes like the fridge, thermostat and fancy lighting, and into heavily regulated industries including aerospace, pharmaceutical/medical devices and energy. “Smartness” will embed itself within individual products that are part of our daily lives. We will engage with smart products - learning from them, informing them, and communicating with them. Smart produc...
Jan. 19, 2017 11:45 AM EST Reads: 1,686
"What is the next step in the evolution of IoT systems? The answer is data, information, which is a radical shift from assets, from things to input for decision making," stated Michael Minkevich, VP of Technology Services at Luxoft, in this SYS-CON.tv interview at @ThingsExpo, held November 3-5, 2015, at the Santa Clara Convention Center in Santa Clara, CA.
Jan. 19, 2017 10:00 AM EST Reads: 5,545
The emerging Internet of Everything creates tremendous new opportunities for customer engagement and business model innovation. However, enterprises must overcome a number of critical challenges to bring these new solutions to market. In his session at @ThingsExpo, Michael Martin, CTO/CIO at nfrastructure, outlined these key challenges and recommended approaches for overcoming them to achieve speed and agility in the design, development and implementation of Internet of Everything solutions with...
Jan. 19, 2017 10:00 AM EST Reads: 5,543
WebRTC sits at the intersection between VoIP and the Web. As such, it poses some interesting challenges for those developing services on top of it, but also for those who need to test and monitor these services. In his session at WebRTC Summit, Tsahi Levent-Levi, co-founder of testRTC, reviewed the various challenges posed by WebRTC when it comes to testing and monitoring and on ways to overcome them.
Jan. 19, 2017 09:30 AM EST Reads: 6,015
Internet of @ThingsExpo, taking place June 6-8, 2017 at the Javits Center in New York City, New York, is co-located with the 20th International Cloud Expo and will feature technical sessions from a rock star conference faculty and the leading industry players in the world. @ThingsExpo New York Call for Papers is now open.
Jan. 19, 2017 07:30 AM EST Reads: 3,628
Smart Cities are here to stay, but for their promise to be delivered, the data they produce must not be put in new siloes. In his session at @ThingsExpo, Mathias Herberts, Co-founder and CTO of Cityzen Data, discussed the best practices that will ensure a successful smart city journey.
Jan. 19, 2017 07:00 AM EST Reads: 2,005
Every successful software product evolves from an idea to an enterprise system. Notably, the same way is passed by the product owner's company. In his session at 20th Cloud Expo, Oleg Lola, CEO of MobiDev, will provide a generalized overview of the evolution of a software product, the product owner, the needs that arise at various stages of this process, and the value brought by a software development partner to the product owner as a response to these needs.
Jan. 19, 2017 07:00 AM EST Reads: 1,209
In 2014, Amazon announced a new form of compute called Lambda. We didn't know it at the time, but this represented a fundamental shift in what we expect from cloud computing. Now, all of the major cloud computing vendors want to take part in this disruptive technology. In his session at 20th Cloud Expo, John Jelinek IV, a web developer at Linux Academy, will discuss why major players like AWS, Microsoft Azure, IBM Bluemix, and Google Cloud Platform are all trying to sidestep VMs and containers...
Jan. 19, 2017 06:00 AM EST Reads: 844
@ThingsExpo has been named the ‘Top WebRTC Influencer' by iTrend. iTrend processes millions of conversations, tweets, interactions, news articles, press releases, blog posts - and extract meaning form them and analyzes mobile and desktop software platforms used to communicate, various metadata (such as geo location), and automation tools. In overall placement, @ThingsExpo ranked as the number one ‘WebRTC Influencer' followed by @DevOpsSummit at 55th.
Jan. 19, 2017 06:00 AM EST Reads: 4,756
SYS-CON Events announced today that MobiDev, a client-oriented software development company, will exhibit at SYS-CON's 20th International Cloud Expo®, which will take place June 6-8, 2017, at the Javits Center in New York City, NY, and the 21st International Cloud Expo®, which will take place October 31-November 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA. MobiDev is a software company that develops and delivers turn-key mobile apps, websites, web services, and complex softw...
Jan. 19, 2017 05:30 AM EST Reads: 1,901
The cloud market growth today is largely in public clouds. While there is a lot of spend in IT departments in virtualization, these aren’t yet translating into a true “cloud” experience within the enterprise. What is stopping the growth of the “private cloud” market? In his general session at 18th Cloud Expo, Nara Rajagopalan, CEO of Accelerite, explored the challenges in deploying, managing, and getting adoption for a private cloud within an enterprise. What are the key differences between wh...
Jan. 19, 2017 01:15 AM EST Reads: 6,111
"Tintri was started in 2008 with the express purpose of building a storage appliance that is ideal for virtualized environments. We support a lot of different hypervisor platforms from VMware to OpenStack to Hyper-V," explained Dan Florea, Director of Product Management at Tintri, in this SYS-CON.tv interview at 18th Cloud Expo, held June 7-9, 2016, at the Javits Center in New York City, NY.
Jan. 19, 2017 12:45 AM EST Reads: 4,728
The security needs of IoT environments require a strong, proven approach to maintain security, trust and privacy in their ecosystem. Assurance and protection of device identity, secure data encryption and authentication are the key security challenges organizations are trying to address when integrating IoT devices. This holds true for IoT applications in a wide range of industries, for example, healthcare, consumer devices, and manufacturing. In his session at @ThingsExpo, Lancen LaChance, vic...
Jan. 18, 2017 09:45 PM EST Reads: 6,535
WebRTC has had a real tough three or four years, and so have those working with it. Only a few short years ago, the development world were excited about WebRTC and proclaiming how awesome it was. You might have played with the technology a couple of years ago, only to find the extra infrastructure requirements were painful to implement and poorly documented. This probably left a bitter taste in your mouth, especially when things went wrong.
Jan. 18, 2017 09:30 PM EST Reads: 7,646
Big Data, cloud, analytics, contextual information, wearable tech, sensors, mobility, and WebRTC: together, these advances have created a perfect storm of technologies that are disrupting and transforming classic communications models and ecosystems. In his session at @ThingsExpo, Erik Perotti, Senior Manager of New Ventures on Plantronics’ Innovation team, provided an overview of this technological shift, including associated business and consumer communications impacts, and opportunities it m...
Jan. 18, 2017 09:30 PM EST Reads: 5,754
You have great SaaS business app ideas. You want to turn your idea quickly into a functional and engaging proof of concept. You need to be able to modify it to meet customers' needs, and you need to deliver a complete and secure SaaS application. How could you achieve all the above and yet avoid unforeseen IT requirements that add unnecessary cost and complexity? You also want your app to be responsive in any device at any time. In his session at 19th Cloud Expo, Mark Allen, General Manager of...
Jan. 18, 2017 07:30 PM EST Reads: 3,161
WebRTC is bringing significant change to the communications landscape that will bridge the worlds of web and telephony, making the Internet the new standard for communications. Cloud9 took the road less traveled and used WebRTC to create a downloadable enterprise-grade communications platform that is changing the communication dynamic in the financial sector. In his session at @ThingsExpo, Leo Papadopoulos, CTO of Cloud9, discussed the importance of WebRTC and how it enables companies to focus o...
Jan. 18, 2017 06:15 PM EST Reads: 4,211
Big Data engines are powering a lot of service businesses right now. Data is collected from users from wearable technologies, web behaviors, purchase behavior as well as several arbitrary data points we’d never think of. The demand for faster and bigger engines to crunch and serve up the data to services is growing exponentially. You see a LOT of correlation between “Cloud” and “Big Data” but on Big Data and “Hybrid,” where hybrid hosting is the sanest approach to the Big Data Infrastructure pro...
Jan. 18, 2017 05:30 PM EST Reads: 4,916
In his General Session at 16th Cloud Expo, David Shacochis, host of The Hybrid IT Files podcast and Vice President at CenturyLink, investigated three key trends of the “gigabit economy" though the story of a Fortune 500 communications company in transformation. Narrating how multi-modal hybrid IT, service automation, and agile delivery all intersect, he will cover the role of storytelling and empathy in achieving strategic alignment between the enterprise and its information technology.
Jan. 18, 2017 04:45 PM EST Reads: 4,626
Buzzword alert: Microservices and IoT at a DevOps conference? What could possibly go wrong? In this Power Panel at DevOps Summit, moderated by Jason Bloomberg, the leading expert on architecting agility for the enterprise and president of Intellyx, panelists peeled away the buzz and discuss the important architectural principles behind implementing IoT solutions for the enterprise. As remote IoT devices and sensors become increasingly intelligent, they become part of our distributed cloud enviro...
Jan. 18, 2017 04:30 PM EST Reads: 4,841