Welcome!

Java Authors: Yeshim Deniz, Liz McMillan, Carmen Gonzalez, Yakov Fain, Gil Allouche

Related Topics: Big Data Journal, Java, AJAX & REA, Cloud Expo, Apache, SDN Journal

Big Data Journal: Article

So What? – Monitoring Hadoop Beyond Ganglia

Don’t just run Hadoop jobs at scale, run them efficiently and at scale

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?

The Basics
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

Conclusion
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!

More Stories By Michael Kopp

Michael Kopp has over 12 years of experience as an architect and developer in the Enterprise Java space. Before coming to CompuwareAPM dynaTrace he was the Chief Architect at GoldenSource, a major player in the EDM space. In 2009 he joined dynaTrace as a technology strategist in the center of excellence. He specializes application performance management in large scale production environments with special focus on virtualized and cloud environments. His current focus is how to effectively leverage BigData Solutions and how these technologies impact and change the application landscape.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


@ThingsExpo Stories
The Internet of Things (IoT) is making everything it touches smarter – smart devices, smart cars and smart cities. And lucky us, we’re just beginning to reap the benefits as we work toward a networked society. However, this technology-driven innovation is impacting more than just individuals. The IoT has an environmental impact as well, which brings us to the theme of this month’s #IoTuesday Twitter chat. The ability to remove inefficiencies through connected objects is driving change throughout every sector, including waste management. BigBelly Solar, located just outside of Boston, is trans...
SYS-CON Events announced today that Matrix.org has been named “Silver Sponsor” of Internet of @ThingsExpo, which will take place on November 4–6, 2014, at the Santa Clara Convention Center in Santa Clara, CA. Matrix is an ambitious new open standard for open, distributed, real-time communication over IP. It defines a new approach for interoperable Instant Messaging and VoIP based on pragmatic HTTP APIs and WebRTC, and provides open source reference implementations to showcase and bootstrap the new standard. Our focus is on simplicity, security, and supporting the fullest feature set.
Predicted by Gartner to add $1.9 trillion to the global economy by 2020, the Internet of Everything (IoE) is based on the idea that devices, systems and services will connect in simple, transparent ways, enabling seamless interactions among devices across brands and sectors. As this vision unfolds, it is clear that no single company can accomplish the level of interoperability required to support the horizontal aspects of the IoE. The AllSeen Alliance, announced in December 2013, was formed with the goal to advance IoE adoption and innovation in the connected home, healthcare, education, aut...
SYS-CON Events announced today that Red Hat, the world's leading provider of open source solutions, will exhibit at Internet of @ThingsExpo, which will take place on November 4–6, 2014, at the Santa Clara Convention Center in Santa Clara, CA. Red Hat is the world's leading provider of open source software solutions, using a community-powered approach to reliable and high-performing cloud, Linux, middleware, storage and virtualization technologies. Red Hat also offers award-winning support, training, and consulting services. As the connective hub in a global network of enterprises, partners, a...

SUNNYVALE, Calif., Oct. 20, 2014 /PRNewswire/ -- Spansion Inc. (NYSE: CODE), a global leader in embedded systems, today added 96 new products to the Spansion® FM4 Family of flexible microcontrollers (MCUs). Based on the ARM® Cortex®-M4F core, the new MCUs boast a 200 MHz operating frequency and support a diverse set of on-chip peripherals for enhanced human machine interfaces (HMIs) and machine-to-machine (M2M) communications. The rich set of periphera...

The only place to be June 9-11 is Cloud Expo & @ThingsExpo 2015 East at the Javits Center in New York City. Join us there as delegates from all over the world come to listen to and engage with speakers & sponsors from the leading Cloud Computing, IoT & Big Data companies. Cloud Expo & @ThingsExpo are the leading events covering the booming market of Cloud Computing, IoT & Big Data for the enterprise. Speakers from all over the world will be hand-picked for their ability to explore the economic strategies that utility/cloud computing provides. Whether public, private, or in a hybrid form, clo...
Software AG helps organizations transform into Digital Enterprises, so they can differentiate from competitors and better engage customers, partners and employees. Using the Software AG Suite, companies can close the gap between business and IT to create digital systems of differentiation that drive front-line agility. We offer four on-ramps to the Digital Enterprise: alignment through collaborative process analysis; transformation through portfolio management; agility through process automation and integration; and visibility through intelligent business operations and big data.
The Transparent Cloud-computing Consortium (abbreviation: T-Cloud Consortium) will conduct research activities into changes in the computing model as a result of collaboration between "device" and "cloud" and the creation of new value and markets through organic data processing High speed and high quality networks, and dramatic improvements in computer processing capabilities, have greatly changed the nature of applications and made the storing and processing of data on the network commonplace.
Be Among the First 100 to Attend & Receive a Smart Beacon. The Physical Web is an open web project within the Chrome team at Google. Scott Jenson leads a team that is working to leverage the scalability and openness of the web to talk to smart devices. The Physical Web uses bluetooth low energy beacons to broadcast an URL wirelessly using an open protocol. Nearby devices can find all URLs in the room, rank them and let the user pick one from a list. Each device is, in effect, a gateway to a web page. This unlocks entirely new use cases so devices can offer tiny bits of information or simple i...
The Internet of Things (IoT) is going to require a new way of thinking and of developing software for speed, security and innovation. This requires IT leaders to balance business as usual while anticipating for the next market and technology trends. Cloud provides the right IT asset portfolio to help today’s IT leaders manage the old and prepare for the new. Today the cloud conversation is evolving from private and public to hybrid. This session will provide use cases and insights to reinforce the value of the network in helping organizations to maximize their company’s cloud experience.
Things are being built upon cloud foundations to transform organizations. This CEO Power Panel at 15th Cloud Expo, moderated by Roger Strukhoff, Cloud Expo and @ThingsExpo conference chair, will address the big issues involving these technologies and, more important, the results they will achieve. How important are public, private, and hybrid cloud to the enterprise? How does one define Big Data? And how is the IoT tying all this together?
TechCrunch reported that "Berlin-based relayr, maker of the WunderBar, an Internet of Things (IoT) hardware dev kit which resembles a chunky chocolate bar, has closed a $2.3 million seed round, from unnamed U.S. and Switzerland-based investors. The startup had previously raised a €250,000 friend and family round, and had been on track to close a €500,000 seed earlier this year — but received a higher funding offer from a different set of investors, which is the $2.3M round it’s reporting."
The Industrial Internet revolution is now underway, enabled by connected machines and billions of devices that communicate and collaborate. The massive amounts of Big Data requiring real-time analysis is flooding legacy IT systems and giving way to cloud environments that can handle the unpredictable workloads. Yet many barriers remain until we can fully realize the opportunities and benefits from the convergence of machines and devices with Big Data and the cloud, including interoperability, data security and privacy.
All major researchers estimate there will be tens of billions devices - computers, smartphones, tablets, and sensors - connected to the Internet by 2020. This number will continue to grow at a rapid pace for the next several decades. Over the summer Gartner released its much anticipated annual Hype Cycle report and the big news is that Internet of Things has now replaced Big Data as the most hyped technology. Indeed, we're hearing more and more about this fascinating new technological paradigm. Every other IT news item seems to be about IoT and its implications on the future of digital busines...
Cultural, regulatory, environmental, political and economic (CREPE) conditions over the past decade are creating cross-industry solution spaces that require processes and technologies from both the Internet of Things (IoT), and Data Management and Analytics (DMA). These solution spaces are evolving into Sensor Analytics Ecosystems (SAE) that represent significant new opportunities for organizations of all types. Public Utilities throughout the world, providing electricity, natural gas and water, are pursuing SmartGrid initiatives that represent one of the more mature examples of SAE. We have s...
The Internet of Things needs an entirely new security model, or does it? Can we save some old and tested controls for the latest emerging and different technology environments? In his session at Internet of @ThingsExpo, Davi Ottenheimer, EMC Senior Director of Trust, will review hands-on lessons with IoT devices and reveal privacy options and a new risk balance you might not expect.
IoT is still a vague buzzword for many people. In his session at Internet of @ThingsExpo, Mike Kavis, Vice President & Principal Cloud Architect at Cloud Technology Partners, will discuss the business value of IoT that goes far beyond the general public's perception that IoT is all about wearables and home consumer services. The presentation will also discuss how IoT is perceived by investors and how venture capitalist access this space. Other topics to discuss are barriers to success, what is new, what is old, and what the future may hold.
Swiss innovators dizmo Inc. launches its ground-breaking software, which turns any digital surface into an immersive platform. The dizmo platform seamlessly connects digital and physical objects in the home and at the workplace. Dizmo breaks down traditional boundaries between device, operating systems, apps and software, transforming the way users work, play and live. It supports orchestration and collaboration in an unparalleled way enabling any data to instantaneously be accessed on any surface, anywhere and made interactive. Dizmo brings fantasies as seen in Sci-fi movies such as Iro...
There’s Big Data, then there’s really Big Data from the Internet of Things. IoT is evolving to include many data possibilities like new types of event, log and network data. The volumes are enormous, generating tens of billions of logs per day, which raise data challenges. Early IoT deployments are relying heavily on both the cloud and managed service providers to navigate these challenges. In her session at 6th Big Data Expo®, Hannah Smalltree, Director at Treasure Data, to discuss how IoT, Big Data and deployments are processing massive data volumes from wearables, utilities and other mach...
This Internet of Nouns trend is still in the early stages and many of our already connected gadgets do provide human benefits over the typical infotainment. Internet of Things or IoT. You know, where everyday objects have software, chips, and sensors to capture data and report back. Household items like refrigerators, toilets and thermostats along with clothing, cars and soon, the entire home will be connected. Many of these devices provide actionable data - or just fun entertainment - so people can make decisions about whatever is being monitored. It can also help save lives.