We’ve reached the 8th installment of our blog series on Cisco’s Big Data and Analytics vision (beginning with Scott Ciccone’s blog on September 23). No doubt by now you have either seen or heard about Cisco’s broad data and analytics portfolio presented at Strataconf in New York on Oct. 15. And if you missed our October 21st executive webcast ‘Unlock Your Competitive Edge with Cisco Big Data and Analytics Solutions,’ please check it out. Now you’re probably eager to know how to make the most of our approach to data analytics. How can you benefit the most—and the most quickly—from data analysis in your organization?
Customers come to us to ask for support in extracting valuable and actionable business insights from their large stocks of network data. Their goal is always to drive both operational efficiencies and new revenue opportunities. Rapid changes in the business environment increase pressure on time-to-value: savings and revenues need to be brought in as quickly as possible. But traditional ways to extract value from data, complicated by volume, velocity and variety issues, often have a very long time-to-value. In fact, data analytics consulting projects historically take a year or longer to complete. Customers get handed large scale implementation plans and, by the time the program is implemented, the wind has changed: the market opportunity has closed, and the business has moved on.
That’s why for some time now I’ve been a student of accelerating time to value for data analytics. Our job is not just to show our customers the hidden business value of their data, but also to bring that value to them fast. We have developed a rapid prototyping, iterative approach that continuously develops actionable insights from network and other sources of data. Our approach contains four steps to help our customers quickly develop, test, and implement business ideas and processes:
Step One: We start by working with customers and identify key use cases through an “Internet of Everything” iterative planning approach. Our experts don’t just present an idea, but a complete, ready-to-test hypothesis, using visualization techniques and an analytics design approach to discover new ways to do business, based on analytics insights.
Step Two: We use a rapid data extraction approach to capture the data needed to test that hypothesis. We fully leverage Cisco’s Connected Analytics platform, enabling automated data collection and simple correlations exploration.
Step Three: Once we have the data we need, we apply a data science approach to build an “analytics sandbox” in which we test the proposed use cases and measure its outcomes. We use rapid prototyping to test theories, quickly working through iterations to develop a truly working business model for our customers’ unique situations. In the process we are able to identify new insights that became the basis for the next use cases.
Step Four: The result is a set of modular Business Insights, which we interpret and thoroughly test, and turn into an actionable plan that we execute. This makes it relatively easy for our experts to integrate insights and actions into our customers’ transformation initiatives—and in a fraction of the time of traditional data-driven solutions.
The world of top down, outside-in consulting, where value comes from individuals’ experience, is gone. Value today is enabled by the capability of companies like Cisco to extract and interpret data about our customers’ core business, enabling agile decision making and rapid process transformation.
As the Internet of Everything becomes a pervasive reality, we see that analytics is what creates value from all of these connections value. To learn more about Cisco’s vision for the Internet of Everything, read Joseph Bradley’s blog on Thursday, October 23! #UnlockBigData
Tags: analytics, Big Data, consulting, Internet of Everything, unlock big data
One hundred is one of the most ubiquitous numbers in life. We see it on money, best-of lists, memorials, compilations, scoring systems, and many other benchmarks. Cisco Unified Computing System™ reached this significant milestone as Cisco UCS claimed its 100th world-record performance result with the fastest 2-socket server on the SPECjbb®2013 MultiJVM benchmark for the max-jOPS metric.
A Cisco UCS® C220 M4 Rack Server powered by the Intel® Xeon® processor E5-2600 v3 family delivered four times the max-jOPS rate that Cisco measured using previous processor generations to set a new world record. This dramatically increased performance of 195,119 max-jOPS is the top 2-socket MultiJVM score for maximum Java operations (max-jOPS).
This world-record of 195,119 max-jOPS is more than three times better than Cisco UCS C240 M3 with Intel Xeon processor E5 v2 family-based result from just 10 months ago, and more than four times better than our Intel Xeon processor E5 family –based result from 18 months ago. This consistent record-setting performance from Cisco UCS blade and rack servers ensures that Cisco will stay ahead of competitors in delivering high performance for Java virtual machines (JVMs) and throughput-intensive Java applications.
The JVM instances ran on a Cisco UCS C220 M4 Rack Server powered by two 18- core Intel Xeon processor E5-2699 v3 CPUs running Red Hat Enterprise Linux (RHEL)Server 6.5 and Oracle Java HotSpot 64-Bit Server Virtual Machine (VM) Version 1.8.0_20. Check out the Performance Brief and the detailed official benchmark disclosure report for additional information on the benchmark configuration.
Let’s see, what does this latest result mean for our customers? This result proves that IT departments that deploy Java applications on Cisco UCS can deliver more throughput and support more users while reducing the complexity of the data center. For customers assessing infrastructure for Java applications, this result demonstrates Cisco’s capability to consistently deliver record-setting performance with every new generation of processor.
It is interesting to note that although all vendors have access to same Intel processors, only Cisco UCS unleashes their power to deliver high performance to applications through the power of unification. The unique, fabric-centric architecture of Cisco UCS integrates the Intel Xeon processors into a system with a better balance of resources that brings processor power to life. . For additional information on Cisco UCS and Cisco UCS Integrated Infrastructure solutions please visit Cisco Unified Computing & Servers web page.
SPEC, and SPECjbb, is registered trademarks of Standard Performance Evaluation Corporation. The benchmark results used to establish world-record status are based on those available at http://www.spec.org as of October 6, 2014.
Tags: Cisco UCS, Cisco UCS Performance Benchmarks, World-record performance
Today’s guest post comes from Ralph Castain, a principle engineer at Intel. The bulk of this post is an email he sent explaining the concept of a “slot” in typical HPC schedulers. This is a little departure from the normal fare on this blog, but is still a critical concept to understand for running HPC applications efficiently. With his permission, I re-publish Ralph’s email here because it’s a great analogy to explain the “slot” concept, which is broadly applicable to HPC users.
The question of “what is a [scheduler] slot” when discussing schedulers came up yesterday and was an obvious source of confusion, so let me try to explain the concept using a simple model.
Suppose I own a fleet of cars at several locations around the country. I am in the business of providing rides for people. Each car has 5 seats in it.
In one location, my clientele doesn’t have much sense of personal space and is willing to be a little crowded. In that location, I sell tickets to share a car, and allow up to 10 people who are going in roughly the same direction to share a single vehicle (hey, they are willing to sit on each other’s lap!).
In another location, my clients aren¹t quite as “friendly” and really prefer to have their own seat. However, they are willing to share the car with others headed in the same direction, so I sell only as many tickets as I have seats -- in this case, up to 5 tickets for a given car.
In a third location, my clients tend to be a little on the large side -- when I have a large passenger, I find that everyone is happier if I don’t fill the backseat. So when a customer flags that they are a little larger than average, I only sell 4 tickets for that car -- i.e., I require that the middle seat in the back be empty so the passengers can spread out a bit. This may require that I schedule that larger client on a different (perhaps larger) car if I already have 4 people for one that would otherwise be available.
In all of the above locations, I will sell another ticket and allow a passenger to enter a car once someone is dropped off. So I try to keep my cars as full as possible by constantly adding a replacement customer when one leaves. However, I never allow more passengers in the car then what that location will tolerate -- if someone tries to give a “free” lift to a person at the side of the road, I block them from doing so. In addition, if someone calls and asks for 8 tickets, I will schedule them across multiple cars according to the local policy.
In yet another location, I have very exclusive customers -- they don’t want anyone in the car with them. In this case, I simply lease them the entire car for the requested duration. They are free to do whatever they want with the vehicle (including picking up as many passengers as they like), so long as they return it clean and in good working condition.
The concept of the “slot” in schedulers is based on that max payload I define for each location. The scheduler is selling “tickets” to the servers/nodes based on some limit set by the system admin, which is usually based on the needs and policies of the local installation. As the above illustration shows, the definition of the “max payload” for a node can vary by site and node, and the scheduler takes into account a variety of requirements when allocating slots to a user.
Once we have an allocation, we then have to assign seats to individual customers. This is the “mapping” policy. When I map processes “by slot,” what I mean is that I start with the first seat in the first car, and assign customers to seats in a sequential fashion, filling all the allocated seats in the first car before starting to fill the second one. This is best for a “chatty” group of customers, but can lead to one car being more heavily loaded than the others.
When I map processes “by node,” I assign the first customer to the first seat in the first car, I assign the next customer to the first allocated seat in the second car, continuing round-robin until all the customers have been assigned an allocated seat. This balances the load in the cars, but is very inefficient if the customers needed to have a conversation.
Obviously, there are lots and lots of ways for allocating and assigning seats within those cars… that’s a whole separate topic. :-)
Tags: HPC, mpi
Carlos Dominguez, Cisco Senior Vice-President in the Office of the Chairman and CEO, will present the luncheon keynote address on October 30 at the Boston Children’s Hospital Global Pediatric Innovation Summit + Awards.
Held over two days, the Boston Children’s Hospital Global Pediatric Innovation Summit + Awards is designed to address unmet needs, solve problems and seize opportunities in pediatric health care. Thought leaders, including clinicians, health care leaders, venture capitalists, policy makers, and payers, will address the most pressing pain points in pediatric medicine.
Dominguez will highlight ground-breaking new trends in technology, including the Internet of Everything, the democratization of knowledge, and exciting changes in the fields of collaboration and communication.
“Groundbreaking technologies tend to develop in siloes – and when they converge, that’s when they become transformative. Healthcare is experiencing that right now. Processing power, analytics, networking, wireless, genomics, and a host of other areas are suddenly coming together and driving profound change and jaw-dropping new possibilities.”
-- Carlos Dominguez, Cisco Senior Vice-President in the Office of the Chairman and CEO
Dominguez participates in a wide array of social media channels and publishes a daily newsletter on futuristic technology, “The Tech Nowist” (http://tinyurl.com/m53oku8). You can find his blogs at http://carlosdominguez.cisco.com and follow him on Twitter at https://twitter.carlosdominguez.
If you are planning to register for the Boston Children’s Hospital Global Pediatric Innovation Summit + Awards, be sure to attend Dominguez’s keynote to learn about recent breakthroughs in pediatric healthcare delivery and the need to embrace technology while creating a culture of innovation.
Carlos Dominguez luncheon keynote
Boston Children’s Hospital Global Pediatric Innovation Summit + Awards
October 30, 2014
The Seaport World Trade Center
Tags: Cisco, healthcare
Cisco VNI Forecast Projects 1.6 Zettabytes of Global IP Traffic by 2018
Last June, Cisco released an update to its annual Visual Networking Index (VNI), 2013 – 2018. The forecast projects a global annual run rate of 1.6 zettabytes IP traffic by 2018 (up from 614 exabytes of global IP traffic in 2013). A zettabyte certainly sounds like a big number (equivalent to 1 billion terabytes, external storage drives generally come in a terabyte denomination). But how can this forecast methodology be applied to an individual cable or a telecommunications provider’s footprint? Our updated Cable/Telco Service Provider Abstract Network (CT-SPAN) tool can help answer this question. CT-SPAN is developed using the latest VNI assumptions for fixed networks and enable you to quickly develop customized service provider scenarios to estimate the annual traffic forecasts and revenue. It’s as easy as 1-2-3.
1. Select your region and service provider type as well as some basic subscriber and network info.
2. Provide some high-level application info.
3. Start reviewing/analyzing your customized subscriber and network projections.
2014 CT-SPAN Enhancements
This year, Read More »
Tags: cable, cable / telco service provider abstract network tool, ct-span tool, Service Provider, telco, telecommunications, visual networking index, vni