|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!
SYS-CON Events announced today that HPM Networks will exhibit at the 17th International Cloud Expo®, which will take place on November 3–5, 2015, at the Santa Clara Convention Center in Santa Clara, CA. For 20 years, HPM Networks has been integrating technology solutions that solve complex business challenges. HPM Networks has designed solutions for both SMB and enterprise customers throughout the San Francisco Bay Area.
Jul. 30, 2015 03:45 PM EDT Reads: 440
MuleSoft has announced the findings of its 2015 Connectivity Benchmark Report on the adoption and business impact of APIs. The findings suggest traditional businesses are quickly evolving into "composable enterprises" built out of hundreds of connected software services, applications and devices. Most are embracing the Internet of Things (IoT) and microservices technologies like Docker. A majority are integrating wearables, like smart watches, and more than half plan to generate revenue with APIs within the next year.
Jul. 30, 2015 02:30 PM EDT
The Internet of Everything (IoE) brings together people, process, data and things to make networked connections more relevant and valuable than ever before – transforming information into knowledge and knowledge into wisdom. IoE creates new capabilities, richer experiences, and unprecedented opportunities to improve business and government operations, decision making and mission support capabilities.
Jul. 30, 2015 01:45 PM EDT Reads: 259
Growth hacking is common for startups to make unheard-of progress in building their business. Career Hacks can help Geek Girls and those who support them (yes, that's you too, Dad!) to excel in this typically male-dominated world. Get ready to learn the facts: Is there a bias against women in the tech / developer communities? Why are women 50% of the workforce, but hold only 24% of the STEM or IT positions? Some beginnings of what to do about it! In her Opening Keynote at 16th Cloud Expo, Sandy Carter, IBM General Manager Cloud Ecosystem and Developers, and a Social Business Evangelist, d...
Jul. 30, 2015 12:00 PM EDT Reads: 2,044
In his keynote at 16th Cloud Expo, Rodney Rogers, CEO of Virtustream, discussed the evolution of the company from inception to its recent acquisition by EMC – including personal insights, lessons learned (and some WTF moments) along the way. Learn how Virtustream’s unique approach of combining the economics and elasticity of the consumer cloud model with proper performance, application automation and security into a platform became a breakout success with enterprise customers and a natural fit for the EMC Federation.
Jul. 30, 2015 09:00 AM EDT Reads: 2,150
The Internet of Things is not only adding billions of sensors and billions of terabytes to the Internet. It is also forcing a fundamental change in the way we envision Information Technology. For the first time, more data is being created by devices at the edge of the Internet rather than from centralized systems. What does this mean for today's IT professional? In this Power Panel at @ThingsExpo, moderated by Conference Chair Roger Strukhoff, panelists addressed this very serious issue of profound change in the industry.
Jul. 29, 2015 03:00 PM EDT Reads: 1,265
Discussions about cloud computing are evolving into discussions about enterprise IT in general. As enterprises increasingly migrate toward their own unique clouds, new issues such as the use of containers and microservices emerge to keep things interesting. In this Power Panel at 16th Cloud Expo, moderated by Conference Chair Roger Strukhoff, panelists addressed the state of cloud computing today, and what enterprise IT professionals need to know about how the latest topics and trends affect their organization.
Jul. 29, 2015 02:00 PM EDT Reads: 1,180
For IoT to grow as quickly as analyst firms’ project, a lot is going to fall on developers to quickly bring applications to market. But the lack of a standard development platform threatens to slow growth and make application development more time consuming and costly, much like we’ve seen in the mobile space. In his session at @ThingsExpo, Mike Weiner, Product Manager of the Omega DevCloud with KORE Telematics Inc., discussed the evolving requirements for developers as IoT matures and conducted a live demonstration of how quickly application development can happen when the need to comply wit...
Jul. 29, 2015 07:30 AM EDT Reads: 281
Explosive growth in connected devices. Enormous amounts of data for collection and analysis. Critical use of data for split-second decision making and actionable information. All three are factors in making the Internet of Things a reality. Yet, any one factor would have an IT organization pondering its infrastructure strategy. How should your organization enhance its IT framework to enable an Internet of Things implementation? In his session at @ThingsExpo, James Kirkland, Red Hat's Chief Architect for the Internet of Things and Intelligent Systems, described how to revolutionize your archit...
Jul. 28, 2015 06:30 PM EDT Reads: 1,378
It is one thing to build single industrial IoT applications, but what will it take to build the Smart Cities and truly society-changing applications of the future? The technology won’t be the problem, it will be the number of parties that need to work together and be aligned in their motivation to succeed. In his session at @ThingsExpo, Jason Mondanaro, Director, Product Management at Metanga, discussed how you can plan to cooperate, partner, and form lasting all-star teams to change the world and it starts with business models and monetization strategies.
Jul. 28, 2015 04:30 PM EDT Reads: 1,760
Converging digital disruptions is creating a major sea change - Cisco calls this the Internet of Everything (IoE). IoE is the network connection of People, Process, Data and Things, fueled by Cloud, Mobile, Social, Analytics and Security, and it represents a $19Trillion value-at-stake over the next 10 years. In her keynote at @ThingsExpo, Manjula Talreja, VP of Cisco Consulting Services, discussed IoE and the enormous opportunities it provides to public and private firms alike. She will share what businesses must do to thrive in the IoE economy, citing examples from several industry sectors.
Jul. 28, 2015 11:00 AM EDT Reads: 2,037
There will be 150 billion connected devices by 2020. New digital businesses have already disrupted value chains across every industry. APIs are at the center of the digital business. You need to understand what assets you have that can be exposed digitally, what their digital value chain is, and how to create an effective business model around that value chain to compete in this economy. No enterprise can be complacent and not engage in the digital economy. Learn how to be the disruptor and not the disruptee.
Jul. 27, 2015 10:00 AM EDT Reads: 2,024
Akana has released Envision, an enhanced API analytics platform that helps enterprises mine critical insights across their digital eco-systems, understand their customers and partners and offer value-added personalized services. “In today’s digital economy, data-driven insights are proving to be a key differentiator for businesses. Understanding the data that is being tunneled through their APIs and how it can be used to optimize their business and operations is of paramount importance,” said Alistair Farquharson, CTO of Akana.
Jul. 27, 2015 09:00 AM EDT Reads: 316
Business as usual for IT is evolving into a "Make or Buy" decision on a service-by-service conversation with input from the LOBs. How does your organization move forward with cloud? In his general session at 16th Cloud Expo, Paul Maravei, Regional Sales Manager, Hybrid Cloud and Managed Services at Cisco, discusses how Cisco and its partners offer a market-leading portfolio and ecosystem of cloud infrastructure and application services that allow you to uniquely and securely combine cloud business applications and services across multiple cloud delivery models.
Jul. 27, 2015 08:00 AM EDT Reads: 1,898
The enterprise market will drive IoT device adoption over the next five years. In his session at @ThingsExpo, John Greenough, an analyst at BI Intelligence, division of Business Insider, analyzed how companies will adopt IoT products and the associated cost of adopting those products. John Greenough is the lead analyst covering the Internet of Things for BI Intelligence- Business Insider’s paid research service. Numerous IoT companies have cited his analysis of the IoT. Prior to joining BI Intelligence, he worked analyzing bank technology for Corporate Insight and The Clearing House Payment...
Jul. 26, 2015 09:00 PM EDT Reads: 1,569
"Optimal Design is a technology integration and product development firm that specializes in connecting devices to the cloud," stated Joe Wascow, Co-Founder & CMO of Optimal Design, in this SYS-CON.tv interview at @ThingsExpo, held June 9-11, 2015, at the Javits Center in New York City.
Jul. 25, 2015 02:00 PM EDT Reads: 383
SYS-CON Events announced today that CommVault has been named “Bronze Sponsor” of SYS-CON's 17th International Cloud Expo®, which will take place on November 3–5, 2015, at the Santa Clara Convention Center in Santa Clara, CA. A singular vision – a belief in a better way to address current and future data management needs – guides CommVault in the development of Singular Information Management® solutions for high-performance data protection, universal availability and simplified management of data on complex storage networks. CommVault's exclusive single-platform architecture gives companies unp...
Jul. 25, 2015 01:00 PM EDT Reads: 1,952
Electric Cloud and Arynga have announced a product integration partnership that will bring Continuous Delivery solutions to the automotive Internet-of-Things (IoT) market. The joint solution will help automotive manufacturers, OEMs and system integrators adopt DevOps automation and Continuous Delivery practices that reduce software build and release cycle times within the complex and specific parameters of embedded and IoT software systems.
Jul. 25, 2015 12:15 PM EDT Reads: 456
"ciqada is a combined platform of hardware modules and server products that lets people take their existing devices or new devices and lets them be accessible over the Internet for their users," noted Geoff Engelstein of ciqada, a division of Mars International, in this SYS-CON.tv interview at @ThingsExpo, held June 9-11, 2015, at the Javits Center in New York City.
Jul. 25, 2015 12:00 PM EDT Reads: 1,530
Internet of Things is moving from being a hype to a reality. Experts estimate that internet connected cars will grow to 152 million, while over 100 million internet connected wireless light bulbs and lamps will be operational by 2020. These and many other intriguing statistics highlight the importance of Internet powered devices and how market penetration is going to multiply many times over in the next few years.
Jul. 25, 2015 09:00 AM EDT Reads: 1,484