Monday Feb 03, 2014

Nebulous Analogies

Having done my doctoral research on analogies, including their benefits and misuses, I am often amused by the analogies that crop up in the field of computing.

Starting long ago with von Neumann's "The Computer and the Brain" [1], computing analogies have both illuminated and confused many important IT concepts. The use of the term "memory" for persistent data storage, for example has a fairly clear and beneficial connection to human recall concepts. Less clearly helpful are analogies to thinking and consciousness common in AI research, as well as the current discussions around "cloud" computing. The corresponding characteristics of meteorological clouds and modern distributed computing are imprecise and misleading. It’s not clear how that analogy helps understand the critical technical concepts.

Now we have yet another unhelpful analogy, “Fog Computing” [2]. Attempting to characterize the “Internet of Things” [3] as an all-pervasive, obscuring “mist” explains nothing about the nature of ubiquitous, embedded computing services. And of course, computing vendors will jump at the chance to exploit the latest analogical buzzword to promote their products [4].

It’s difficult to communicate anything without using analogies, since that’s how the human brain works — we think using analogies [5]. But we should be careful in selecting the source analogs when trying to explain complex concepts. Poorly chosen sources can confuse and limit thinking and can hinder solution development. Surely there are better source analogs than clouds and fog. The “web” is clearly better than the “cloud” in conveying the idea of connectivity. But what is the ideal analogy for computing services that will eventually fill every corner of our daily lives, using our always-connected devices like smartphones and tablets, and the embedded services in our homes, cars, businesses, and social media? I don’t think it’s atmospheric phenomena. And I’m not sure it’s the “invisible computer” analogy either [6]. It’s probably more like oxygen [7], although that implies that we can’t live without it.
Hmmm…maybe we can’t.


[1] http://www.amazon.com/Computer-Silliman-Memorial-Lectures-Series/dp/0300181116
[2] http://conferences.sigcomm.org/sigcomm/2012/paper/mcc/p13.pdf
[3] http://en.wikipedia.org/wiki/Internet_of_Things
[4] http://newsroom.cisco.com/release/1334100/Cisco-Delivers-Vision-of-Fog-Computing-to-Accelerate-Value-from-Billions-of-Connected-Devices
[5] http://www.amazon.com/Surfaces-Essences-Analogy-Fuel-Thinking/dp/0465018475
[6] http://www.amazon.com/The-Invisible-Future-Integration-Technology/dp/0071382240
[7] http://oxygen.lcs.mit.edu/Overview.html

Sunday Jan 05, 2014

Repost from Washington Post: 5 Myths About the Cloud

Interesting article in the 5 JAN 2014 Washington Post, 5 Myths About the Cloud. Some funny survey results from the tech-challenged public, and some good points about cloud security, reliability, and environmental impact. And 2014 looks to be a very "cloudy" year, with more companies and government agencies deploying mission critical applications and data in both public and private clouds. And although many potential cloud computing customers still don't think of Oracle as a major player in this growing technology services market, they will if they do their homework.

If you want to play around with your own Hadoop-based cloud, be sure to check out these How-To articles on the Oracle Technology Network:

Hey! You! Get onto my cloud!

Tuesday Nov 05, 2013

Brendan Gregg's "Systems Performance: Enterprise and the Cloud"

An outstanding new text on UNIX & Linux system performance by one of the best experts.[Read More]

Friday Aug 02, 2013

READ_ME_FIRST: What Do I Do With All of Those SPARC Threads?

New Oracle Technical White Paper: READ_ME_FIRST: What Do I Do With All of Those SPARC Threads?

Executive Overview

With an amazing 1,536 threads in an Oracle M5-32 system, the number of threads in a single system has never been so high. This offers a tremendous processing capacity, but one may wonder how to make optimal use of all these resources.

In this technical white paper, we explain how the heavily threaded Oracle T5 and M5 servers can be deployed to efficiently consolidate and manage workloads using virtualization through Oracle Solaris Zones, Oracle VM Server for SPARC, and Oracle Enterprise Manager Ops Center, as well as how to improve the performance of a single application through multi-threading.

READ_ME_FIRST: What Do I Do With All of Those SPARC Threads?

Thursday Jun 27, 2013

Ginormous...the new Oracle T5-8 SPARC SuperCluster!

Ginormous...no other way to describe it...the new Oracle T5-8 SPARC SuperCluster...2000+ fast CPU threads, massive memory (DRAM & Flash), 1.2 M IOPS, HUGE storage and bandwidth...WOW! Wanna build a SPARC Cloud? This is it! Multiple virtualization technologies (VM Server for SPARC, and Solaris zones) for elasticity and resource pooling along with Oracle Enterprise Manager Ops Center 12c providing full cloud management capability.

Check it out!:

Monday Mar 25, 2013

Join Oracle's CEO Larry Ellison & VP John Fowler for a live web event: Tues., Mar. 26 @ 3pm CT

Join Oracle's CEO Larry Ellison & VP John Fowler for a live web event: Tues., Mar. 26 @ 3pm CT Register: HERE

Wednesday Jan 16, 2013

NIST Cloud Computing & Big Data Forum, Jan 15-17 2012

NIST Cloud Computing & Big Data Forum, Jan 15-17 2012

On the second day of the NIST Cloud Computing & Big Data Forum, we had a real treat...the keynote speaker was Vint Cerf, Google's VP and Chief Internet Evangelist. Allocated a mere 30 minutes, he actually spoke for more than 45, detailing his thoughts on Cloud Computing, Big Data, and related topics. He highlighted some of the security issues concerning cloud computing, like the problem of "leftovers" (data remaining after a virtual service has been deprovisioned), and the need for strong authentication of both user identities and trusted identifiers. He also emphasized the need for non-proprietary inter-cloud communication and collaboration protocols, and mentioned a bit about Google's cloud services including a comment about their current 100Gb OpenFlow-based infrastructure and their potential need for Terabit connectivity in their data centers.

One interesting Big Data comment he made concerned the problem of feeding data fast enough into today's powerful multicore processors, and suggested the memristor as a possible technology solution.

Later in the conference at a panel on Big Data Use Cases, Veterans Affairs CTO Peter Levin gave a briefing on the VA's Blue Button system for accessing veterans' health care records, and its potential to grow to thousands of terabytes as genetic data is included, just one of many examples of Big Data projects discussed today. Presentations from the conference will be posted on the NIST Cloud Computing Web site.

Friday Nov 02, 2012

Meet the co-author: Solaris 11, at Oracle OpenWorld 2012

Meet the co-author: Solaris 11, at Oracle OpenWorld 2012:
http://vimeo.com/oraclecert/review/52719812/2edb3f84e7

Sunday Sep 23, 2012

ReBlog: So You Want To Build a SPARC Cloud

In case you missed this on Steve Wilson's Blog:
So You Want To Build a SPARC Cloud

Thursday Sep 06, 2012

The first Oracle Solaris 11 book is now available

The first Oracle Solaris 11 book is now available:

Oracle Solaris 11 System Administration - The Complete Reference
by Michael Jang, Harry Foxwell, Christine Tran, and Alan Formy-Duval

  • The book covers the Oracle Solaris 11 11/11 release; although the next OS release will be available soon, the book covers major topics and features that are not expected to change significantly.
  • The target audience is broad, and includes Solaris admins, Linux admins and developers, and even those somewhat unfamiliar with UNIX.
  • The coauthors include practitioners and developers from outside of Oracle, emphasizing their field experience using Solaris 11.
  • The book complements the extensive Oracle Solaris 11 Information Library, and covers the main system administration topics of installation, configuration, and management.

More Oracle Solaris 11 info here

Friday Aug 31, 2012

New Book: Oracle Exalogic Elastic Cloud Handbook

Oracle Exalogic Elastic Cloud Handbook, by Tom Plunkett, TJ Palazzolo, and Tejas Joshi, Oracle Press.

The well-known characteristics and tiers of cloud computing have spawned myriad implementations by a host of vendors and system integrators. One of these, Oracle's Exalogic Elastic Cloud, part of Oracle's family of Engineered Systems, is a key component of Oracle's public and private cloud computing solutions, providing critical PaaS (Platform as a Service) features for cloud developers. These developers need guidance to take advantage of Exalogic's extensive capabilities, and the Oracle Exalogic Elastic Cloud Handbook, written by three highly experienced Oracle technologists, provides that guidance.

Part One of the book covers Exalogic's hardware and software components, and includes a very useful chapter on deployment examples, describing best practices for scalabiity, availability, backup and recovery, and multi-tenant security, including integration with other Oracle Engineered Systems and products such as Exadata and storage subsystems.

Part Two is a thorough guide to Exalogic installation features, configuration and monitoring, packaged application software management, and scalable application development.

The book also provides an extensive list of online resources, including pointers to Web sites, whitepapers, instructional videos, and other Oracle documentation.

So, if you're planning to implement Exalogic as part of your cloud infrastructure, or are considering such, you'll find lots of sage advice and best practices in this handbook.

Tuesday Aug 28, 2012

The Growing Importance of Network Virtualization

The Growing Importance of Network Virtualization

We often focus on server virtualization when we discuss cloud computing, but just as often we neglect to consider some of the critical implications of that technology. The ability to create virtual environments (or VEs [1]) means that we can create, destroy, activate and deactivate, and more importantly, MOVE them around within the cloud infrastructure. This elasticity and mobility has profound implications for how network services are defined, managed, and used to provide cloud services. It's not just servers that benefit from virtualization, it's the network as well.

Network virtualization is becoming a hot topic, and not just for discussion but for companies like Oracle and others who have recently acquired net virtualization companies [2,3]. But even before this topic became so prominent, Solaris engineers were working on technologies in Solaris 11 to virtualize network services, known as Project Crossbow [4].

And why is network virtualization so important? Because old assumptions about network devices, topology, and management must be re-examined in light of the self-service, elasticity, and resource sharing requirements of cloud computing infrastructures. Static, hierarchical network designs, and inter-system traffic flows, need to be reconsidered and quite likely re-architected to take advantage of new features like virtual NICs and switches, bandwidth control, load balancing, and traffic isolation. For example, traditional multi-tier Web services (Web server, App server, DB server) that share net traffic over Ethernet wires can now be virtualized and hosted on shared-resource systems that communicate within a larger server at system bus speeds, increasing performance and reducing wired network traffic. And virtualized traffic flows can be monitored and adjusted as needed to optimize network performance for dynamically changing cloud workloads. Additionally, as VEs come and go and move around in the cloud, static network configuration methods cannot easily accommodate the routing and addressing flexibility that VE mobility implies; virtualizing the network itself is a requirement.

Oracle Solaris 11 [5] includes key network virtualization technologies needed to implement cloud computing infrastructures. It includes features for the creation and management of virtual NICs and switches, and for the allocation and control of the traffic flows among VEs [6]. Additionally it allows for both sharing and dedication of hardware components to network tasks, such as allocating specific CPUs and vNICs to VEs, and even protocol-specific management of traffic.

So, have a look at your current network topology and management practices in view of evolving cloud computing technologies. And don't simply duplicate the physical architecture of servers and connections in a virtualized environment…rethink the traffic flows among VEs and how they can be optimized using Oracle Solaris 11 and other Oracle products and services.

[1] I use the term "virtual environment" or VE here instead of the more commonly used "virtual machine" or VM, because not all virtualized operating system environments are full OS kernels under the control of a hypervisor…in other words, not all VEs are VMs. In particular, VEs include Oracle Solaris zones, as well as SPARC VMs (previously called LDoms), and x86-based Solaris and Linux VMs running under hypervisors such as OEL, Xen, KVM, or VMware.

[2] Oracle follows VMware into network virtualization space with Xsigo purchase; http://www.mercurynews.com/business/ci_21191001/oracle-follows-vmware-into-network-virtualization-space-xsigo

[3] Oracle Buys Xsigo; http://www.oracle.com/us/corporate/press/1721421

[4] Oracle Solaris 11 Networking Virtualization Technology, http://www.oracle.com/technetwork/server-storage/solaris11/technologies/networkvirtualization-312278.html

[5] Oracle Solaris 11; http://www.oracle.com/us/products/servers-storage/solaris/solaris11/overview/index.html

[6] For example, the Solaris 11 'dladm' command can be used to limit the bandwidth of a virtual NIC, as follows: dladm create-vnic -l net0 -p maxbw=100M vnic0

Wednesday Jun 06, 2012

Clouds Aroud the World

At the NIST Cloud Computing Workshop this week; representatives from Canada, China, and Japan presented on their cloud computing efforts. Some interesting points made:

Canada: Building "Service Canada" cloud for all citizen services, but raised the issue of data location...cloud data must be within Canada border, so they will not focus on public clouds where they don't know or can't control data location.

Japan: In response to the massive destruction of the Great East Japan Earthquake, Japan is building nation-wide cloud services to support disaster relief, data recovery, and support for rebuilding new communities.

US Ambassador Philip Verveer discussed the need for international cooperation and standards development to enable interoperability of cloud services, keeping in mind cultural and political differences. Additionally, an industry panel reported on cloud standards development, including some actual interoperability testing at http://www.cloudplugfest.org. Much of the first two days of the workshop covered progress and action plans around the 10 High-Priority Requirements to Further USG Agency Cloud Computing Adoption.

Thursday's sessions will cover the work of the various NIST Cloud Computing Working Groups on

  • Reference Architecture and Taxonomy
  • Standards Acceleration to Jumpstart the Adoption of Cloud Computing (SAJACC)
  • Cloud Security
  • Standards Roadmap
  • Business Use Cases

(see Working Groups of NIST Cloud Computing )

Tuesday Nov 15, 2011

What's a "Cloud Operating System"?

What's a "Cloud Operating System"?

Oracle's recently introduced Solaris 11 has been touted as "The First Cloud OS". Interesting claim, but what exactly does it mean? To answer that, we need to recall what characteristics define a cloud and then see how Solaris 11's capabilities map to those characteristics.

By now, most cloud computing professionals have at least heard of, if not adopted, the National Institute of Standards and Technology (NIST) Definition of Cloud Computing, including its vocabulary and conceptual architecture. NIST says that cloud computing includes these five characteristics:

  1. On-demand self-service
  2. Broad network access
  3. Resource pooling
  4. Rapid elasticity
  5. Measured service
How does Solaris 11 support these capabilities? Well, one of the key enabling technologies for cloud computing is virtualization, and Solaris 11 along with Oracle's SPARC and x86 hardware offerings provides the full range of virtualization technologies including dynamic hardware domains, hypervisors for both x86 and SPARC systems, and efficient non-hypervisor workload virtualization with containers. This provides the elasticity needed for cloud systems by supporting on-demand creation and resizing of application environments; it supports the safe partitioning of cloud systems into multi-tenant infrastructures, adding resources as needed and deprovisioning computing resources when no longer needed, allowing for pay-only-for-usage chargeback models.

For cloud computing developers, add to that the next generation of Java, and you've got the NIST requirements covered. The results, or one of them anyway, are services like the new Oracle Public Cloud. And Solaris is the ideal platform for running your Java applications.

So, if you want to develop for cloud computing, for IaaS, PaaS, or SaaS, start with an operating system designed to support cloud's key requirements…start with Solaris 11.

Friday Nov 04, 2011

The NIST Cloud Computing Forum & Workshops IV (Nov 2-4, 2011)

The new US CIO, Steve Van Roekel, along with senior researchers at NIST, hosted the fourth Cloud Computing Forum and Workshops this week ( http://www.nist.gov/itl/cloud/cloudworkshopiv.cfm ). One highlight was the release of the Draft Cloud Technology Roadmap with a call for public comments. See the Web site for the agenda and downloadable documents and presentations.

There were international participants at the event, with much friendly discussion of "openness", "interoperability", and an idealistic "One Cloud" vision of a "Cloud Without Borders". A very hopeful perspective, but perhaps a bit overly optimistic one given the current political state of the world and various governments' control of Internet access and resources.

One issue that concerns me in all this rush to cloud computing is the question of where the expertise will come from to design, build, and manage massive cloud infrastructures? Concepts such as parallel programming, scalability, virtualization, and cache management need to be integrated into CS curricula from the start, maybe even starting in high school but certainly at the undergraduate level. I don't yet see sufficient emphasis on those areas in the CS courses and textbooks offered by many universities. Without a continuous stream of knowledgeable graduates, the lack of cloud computing experience and expertise will slow the adoption of this transformative technology.

Security and trust in the cloud remain primary concerns; the NIST Cloud Computing Security Working Group has released a draft publication outlining 17 key requirement areas for cloud security ( http://collaborate.nist.gov/twiki-cloud-computing/pub/CloudComputing/Documents/NIST_Security_Requirements_for_US_Government_Cloud.pdf ). Yet in spite of the current lack of mature security solutions and interoperability/development standards, it was still recommended that agencies start their cloud deployments with the expectation that expertise will evolve through experimentation, trial, and (inevitably) error.

About

The purpose of this blog is to highlight and to explore general issues around "Cloud Computing" -- its benefits, risks, and component technologies -- and how they are evolving. I'll also periodically comment (of course!) on Oracle's Cloud Computing capabilities, resources, and cloud-related events. -- Harry J Foxwell, PhD, Principal Consultant for Cloud Computing, Oracle Public Sector HW

Search

Categories
Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today