high_scalability high_scalability-2010 high_scalability-2010-835 knowledge-graph by maker-knowledge-mining
Source: html
Introduction: How Big is a Yottabyte? Not so big that the NSA can't hope to store it says CrunchGear : There are a thousand gigabytes in a terabyte, a thousand terabytes in a petabyte, a thousand petabytes in an exabyte, a thousand exabytes in a zettabyte, and a thousand zettabytes in a yottabyte. In other words, a yottabyte is 1,000,000,000,000,000GB . The CMS data aggregation system . The Large Hadron Collider project is using MongoDB as a cache . Here we discuss a new data aggregation system which consumes, indexes and delivers information from different relational and non-relational data sources to answer cross data-service queries and explore meta-data associated with petabytes of experimental data. Google I/O 2010 Videos are up available (many of them anyway). You might be particularly interested in Google Storage for Developers , Building high-throughput data pipelines with Google App Engine , Batch data processing with App Engine , BigQuery and Prediction APIs , and Measure
sentIndex sentText sentNum sentScore
1 Not so big that the NSA can't hope to store it says CrunchGear : There are a thousand gigabytes in a terabyte, a thousand terabytes in a petabyte, a thousand petabytes in an exabyte, a thousand exabytes in a zettabyte, and a thousand zettabytes in a yottabyte. [sent-2, score-1.415]
2 Here we discuss a new data aggregation system which consumes, indexes and delivers information from different relational and non-relational data sources to answer cross data-service queries and explore meta-data associated with petabytes of experimental data. [sent-6, score-0.203]
3 You might be particularly interested in Google Storage for Developers , Building high-throughput data pipelines with Google App Engine , Batch data processing with App Engine , BigQuery and Prediction APIs , and Measure in milliseconds redux: Meet Speed Tracer. [sent-8, score-0.077]
4 dbShards is a software product that allows database sharding to be applied to existing applications and databases with little or no modification to existing code . [sent-14, score-0.074]
5 I talked a little with them at Gluecon and attended their talk. [sent-15, score-0.074]
6 The Google App Engine may be having problems, they are pioneering a new development and deployment model after all, but you gotta love how they deal with problems head on: Datastore Performance Growing Pains . [sent-19, score-0.077]
7 Alexey Kovyrin looks at a few techniques we use at Scribd to solve cache invalidation problems . [sent-24, score-0.135]
8 The two biggest changes were to pipeline the page content to overlap generation, network, and render time, and to move to a very small core JavaScript library for features that are required on the initial page load . [sent-27, score-0.083]
9 James Hamilton: CPU intensive and workloads with poor scaling characteristic are poor choices to be hosted on very low-power, low-cost servers. [sent-29, score-0.422]
10 CPU intensive workloads are a lose because these workloads are CPU-bound so run best where there is maximum CPU per-server in the cluster. [sent-30, score-0.394]
11 By contrast, MySQL premium support would cost about $5,000 per year per node, or $250,000 per year--more than double the Cassandra setup, Stump says, and Microsoft SQL Server can cost as much as $55,000 per processor per year . [sent-32, score-0.42]
12 The $10,000 is an operational expense as opposed to a capital expense, and that's "a bit nicer on the books," he says. [sent-33, score-0.199]
13 Has anyone ever witnessed a hash collision in the wild (MD5, SHA, etc)? [sent-34, score-0.169]
14 The first 1969 moon trip, the Lunar Lander had 10,000 lines of code, 36K of ROM, 2K or RAM, and a 43Khz clock cycle. [sent-39, score-0.343]
15 In 2019 the estimate to go to the moon will take 10 million lines of code, 1GB of RAM, and a 1Ghz clock cycle. [sent-40, score-0.343]
16 Sponsored Jobs Huffington Post- Linux System Administrator Get Your High Scalability Fix at Digg Lolapps - We're a young and scrappy startup in the social application space. [sent-47, score-0.089]
wordName wordTfidf (topN-words)
[('thousand', 0.227), ('stump', 0.226), ('scribd', 0.198), ('yottabyte', 0.186), ('workloads', 0.15), ('moon', 0.145), ('invalidation', 0.135), ('expense', 0.116), ('clock', 0.116), ('june', 0.114), ('aggregation', 0.102), ('petabytes', 0.101), ('cassandra', 0.101), ('devloper', 0.099), ('jobshuffington', 0.099), ('redux', 0.099), ('rom', 0.099), ('zettabyte', 0.099), ('zettabytes', 0.099), ('intensive', 0.094), ('poor', 0.089), ('facebookby', 0.089), ('scrub', 0.089), ('northscale', 0.089), ('scrappy', 0.089), ('administratorget', 0.089), ('collider', 0.089), ('stealth', 0.089), ('witnessed', 0.089), ('alexey', 0.085), ('exabyte', 0.085), ('hadron', 0.085), ('per', 0.084), ('overlap', 0.083), ('ingoogle', 0.083), ('nicer', 0.083), ('lines', 0.082), ('collision', 0.08), ('exabytes', 0.08), ('healing', 0.079), ('pioneering', 0.077), ('pipelines', 0.077), ('ta', 0.077), ('nsa', 0.077), ('app', 0.075), ('toadvertisea', 0.075), ('usfor', 0.075), ('modification', 0.074), ('attended', 0.074), ('consumes', 0.074)]
simIndex simValue blogId blogTitle
same-blog 1 1.0000004 835 high scalability-2010-06-03-Hot Scalability Links for June 3, 2010
Introduction: How Big is a Yottabyte? Not so big that the NSA can't hope to store it says CrunchGear : There are a thousand gigabytes in a terabyte, a thousand terabytes in a petabyte, a thousand petabytes in an exabyte, a thousand exabytes in a zettabyte, and a thousand zettabytes in a yottabyte. In other words, a yottabyte is 1,000,000,000,000,000GB . The CMS data aggregation system . The Large Hadron Collider project is using MongoDB as a cache . Here we discuss a new data aggregation system which consumes, indexes and delivers information from different relational and non-relational data sources to answer cross data-service queries and explore meta-data associated with petabytes of experimental data. Google I/O 2010 Videos are up available (many of them anyway). You might be particularly interested in Google Storage for Developers , Building high-throughput data pipelines with Google App Engine , Batch data processing with App Engine , BigQuery and Prediction APIs , and Measure
2 0.14113565 842 high scalability-2010-06-16-Hot Scalability Links for June 16, 2010
Introduction: You're Doing it Wrong by Poul-Henning Kamp. Don't look so guilty, he's not talking about you know what, he's talking about writing high-performance server programs: Not just wrong as in not perfect, but wrong as in wasting half, or more, of your performance. What good is an O(log2(n)) algorithm if those operations cause page faults and slow disk operations? For most relevant datasets an O(n) or even an O(n^2) algorithm, which avoids page faults, will run circles around it. A Microsoft Windows Azure primer: the basics by Peter Bright. Nice article explaining the basics of Azure and how it compares to Google and Amazon. A call to change the name from NoSQL to Postmodern Databases . Interesting idea, but the problem is the same one I have for Postmodern Art, when is it? I always feel like I'm in the post-post modern period, yet for art it's really in the early 1900s. Let's save future developers from this existential time crisis. Constructions from Dots and Lines by M
3 0.12972857 1320 high scalability-2012-09-11-How big is a Petabyte, Exabyte, Zettabyte, or a Yottabyte?
Introduction: This is an intuitive look at large data sizes By Julian Bunn in Globally Interconnected Object Databases . Bytes(8 bits) 0.1 bytes: A binary decision 1 byte: A single character 10 bytes: A single word 100 bytes: A telegram OR A punched card Kilobyte (1000 bytes) 1 Kilobyte: A very short story 2 Kilobytes: A Typewritten page 10 Kilobytes: An encyclopaedic page OR A deck of punched cards 50 Kilobytes: A compressed document image page 100 Kilobytes: A low-resolution photograph 200 Kilobytes: A box of punched cards 500 Kilobytes: A very heavy box of punched cards Megabyte (1 000 000 bytes) 1 Megabyte: A small novel OR A 3.5 inch floppy disk 2 Megabytes: A high resolution photograph 5 Megabytes: The complete works of Shakespeare OR 30 seconds of TV-quality video 10 Megabytes: A minute of high-fidelity sound OR A digital chest X-ray 20 Megabytes: A box of floppy disks 50 Megabytes: A digital mammogram 100 Megabyte
4 0.12704648 806 high scalability-2010-04-08-Hot Scalability Links for April 8, 2010
Introduction: Scalability porn (SFW). Real time meter for the number of ads being served by doubleclick. Amazing. A constant ~390,000 impressions a second are being served and 25 trillion since 1996. Thanks to Mike Rhoads for title idea. Scalability? Don't worry. Application complexity? Worry by Joe McKendrick. The next challenge on enterprise agendas: application complexity. This is something that lots of hardware — whether from the cloud or internal data center — cannot fix Leo Laporte and Steve Gibson talked about how the iPad was a denial of service attack on UPS delivery schedules. UPS trucks were filled with iPads. Cassandra: Fact vs fiction . Jonathan Ellies puts the beatdown on Cassandra misinformation. Don't you dare say Cassandra can't work across datacenters! JIT'd code calling conventions . Cliff Click Jr shows how Java’s calling convention can match compiled C code in speed, but allows for the flexibility of calling (code,slow) non-JIT'd code . Some assembly code re
5 0.12089804 799 high scalability-2010-03-23-Digg: 4000% Performance Increase by Sorting in PHP Rather than MySQL
Introduction: O'Reilly Radar's James Turner conducted a very informative interview with Joe Stump, current CTO of SimpleGeo and former lead architect at Digg , in which Joe makes some of his usually insightful comments on his experience using Cassandra vs MySQL. As Digg started out with a MySQL oriented architecture and has recently been moving full speed to Cassandra, his observations on some of their lessons learned and the motivation for the move are especially valuable. Here are some of the key takeaways you find useful: Precompute on writes, make reads fast . This is an oldie as a scaling strategy, but it's valuable to see how SimpleGeo is applying it to their problem of finding entities within a certain geographical region. Using Cassandra they've built two clusters: one for indexes and one for records. The records cluster, as you might imagine, is a simple data lookup. The index cluster has a carefully constructed key for every lookup scenario. The indexes are computed on the wr
6 0.12084085 901 high scalability-2010-09-16-How Can the Large Hadron Collider Withstand One Petabyte of Data a Second?
7 0.1193962 802 high scalability-2010-04-01-Hot Scalability Links for April 1, 2010
8 0.11425611 811 high scalability-2010-04-16-Hot Scalability Links for April 16, 2010
9 0.11165968 538 high scalability-2009-03-16-Are Cloud Based Memory Architectures the Next Big Thing?
10 0.11151397 750 high scalability-2009-12-16-Building Super Scalable Systems: Blade Runner Meets Autonomic Computing in the Ambient Cloud
12 0.11109166 1626 high scalability-2014-04-04-Stuff The Internet Says On Scalability For April 4th, 2014
13 0.11067659 954 high scalability-2010-12-06-What the heck are you actually using NoSQL for?
14 0.10991226 517 high scalability-2009-02-21-Google AppEngine - A Second Look
15 0.10953444 313 high scalability-2008-05-02-Friends for Sale Architecture - A 300 Million Page View-Month Facebook RoR App
16 0.1070813 1240 high scalability-2012-05-07-Startups are Creating a New System of the World for IT
17 0.1064942 360 high scalability-2008-08-04-A Bunch of Great Strategies for Using Memcached and MySQL Better Together
18 0.10647684 920 high scalability-2010-10-15-Troubles with Sharding - What can we learn from the Foursquare Incident?
19 0.10539228 721 high scalability-2009-10-13-Why are Facebook, Digg, and Twitter so hard to scale?
20 0.10473522 827 high scalability-2010-05-14-Hot Scalability Links for May 14, 2010
topicId topicWeight
[(0, 0.226), (1, 0.092), (2, 0.015), (3, -0.01), (4, 0.033), (5, 0.032), (6, -0.054), (7, 0.009), (8, 0.029), (9, 0.021), (10, 0.026), (11, -0.053), (12, 0.017), (13, 0.053), (14, -0.065), (15, -0.0), (16, -0.047), (17, -0.001), (18, 0.032), (19, 0.031), (20, 0.029), (21, 0.03), (22, 0.079), (23, -0.037), (24, 0.018), (25, 0.008), (26, 0.008), (27, -0.014), (28, -0.041), (29, 0.016), (30, -0.024), (31, 0.004), (32, 0.018), (33, -0.009), (34, 0.011), (35, -0.024), (36, -0.007), (37, 0.043), (38, 0.029), (39, 0.021), (40, -0.007), (41, -0.003), (42, -0.002), (43, 0.011), (44, 0.013), (45, -0.008), (46, -0.041), (47, 0.025), (48, 0.006), (49, -0.026)]
simIndex simValue blogId blogTitle
same-blog 1 0.96853298 835 high scalability-2010-06-03-Hot Scalability Links for June 3, 2010
Introduction: How Big is a Yottabyte? Not so big that the NSA can't hope to store it says CrunchGear : There are a thousand gigabytes in a terabyte, a thousand terabytes in a petabyte, a thousand petabytes in an exabyte, a thousand exabytes in a zettabyte, and a thousand zettabytes in a yottabyte. In other words, a yottabyte is 1,000,000,000,000,000GB . The CMS data aggregation system . The Large Hadron Collider project is using MongoDB as a cache . Here we discuss a new data aggregation system which consumes, indexes and delivers information from different relational and non-relational data sources to answer cross data-service queries and explore meta-data associated with petabytes of experimental data. Google I/O 2010 Videos are up available (many of them anyway). You might be particularly interested in Google Storage for Developers , Building high-throughput data pipelines with Google App Engine , Batch data processing with App Engine , BigQuery and Prediction APIs , and Measure
2 0.83494127 1472 high scalability-2013-06-07-Stuff The Internet Says On Scalability For June 7, 2013
Introduction: Hey, it's HighScalability time: ( Ever feel like everyone has already climbed your Everest?) Trillion Particles, 120,000 cores, and 350 TBs: Lessons Learned From a Hero I/O Run on Hopper Quotable Quotes: @PenLlawen : @spolsky In my time as a scalability engineer, I’ve seen plenty of cases where optimisation was left too late. Even harder to fix. @davidlubar : Whoever said you can't fold a piece of paper in half more than 7 times probably forget to unfold it each time. I'm up to 6,000. deno : A quick comparison of App Engine vs. Compute Engine prices shows that App Engine is at best 10x more expensive per unit of RAM. Fred Wilson : strategy is figuring out what part of the market the company wants to play in, how it goes to market, and how it differentiates itself in the market it is about what you are going to do and importantly what you are not going to do Elon Musk : SpaceX was able to achieve orders of magnitude saving
3 0.81252718 1080 high scalability-2011-07-15-Stuff The Internet Says On Scalability For July 15, 2011
Introduction: Submitted for your scaling pleasure: That's a lot of data...CERN: ATLAS produces up to 320M bytes per second, followed by CMS with 220M Bps . Amazon Cloud Now Stores 339 Billion Objects . CERN also has an open source hardware effort. Domas Mituzas on why Facebook may just outlast their MySQL heritage: I feel somewhat sad that I have to put this truism out here: disks are way more cost efficient, and if used properly can be used to facilitate way more long-term products, not just real time data. Think Wikipedia without history, think comments that disappear on old posts, together with old posts, think all 404s you hit on various articles you remember from the past and want to read. Building the web that lasts is completely different task from what academia people imagine building the web is. What happens in real world if one gets 2x efficiency gain? Twice more data can be stored, twice more data intensive products can be launched. Quotes that are quoted because they are
4 0.81044769 1424 high scalability-2013-03-15-Stuff The Internet Says On Scalability For March 15, 2013
Introduction: Hey, it's HighScalability time: 0: # of Google Readers; 2.5 billion /day: new pieces of content added to Facebook; 2.7 billion /day: likes added to Facebook; 7PB/month : photos added to Facebook Quotable Quotes: @cwgem : It seems like cutting down API access is the stock scalability answer these days @abenik : @Prismatic surfaced this article on their architecture for me. How meta. @NewsBlur : The waters are rocky now, but take note that I have some time to get things right. I'm working this week to get things stable, then scale. @Pinboard : Just learned that Google Reader no longer offers direct JSON export. I guess they held the annual "What should we ruin next?" staff retreat @DEVOPS_BORAT : You can not able have unlimit scalability without unlimit outage. Jeff : Amazon RDS Scales Up - Provision 3 TB and 30,000 IOPS Per DB Instance @migueldeicaza : Google recently hired al
5 0.80390221 1499 high scalability-2013-08-09-Stuff The Internet Says On Scalability For August 9, 2013
Introduction: Hey, it's HighScalability time: 25% : Percentage of North American Internet Traffic served by Google Quotable Quotes: Aristotle: We must not expect more precision than the subject-matter admits. Bret Victor : Technologies change quickly and minds change slowly. Ideas require people to unlearn what they’ve learned and adopt new ideas. They think what they’ve learned is programming and this new stuff isn’t programming. Steven Roberts: Art without engineering is dreaming; engineering without art is calculating. @jamesurquhart : “@b6n: @jamesurquhart @krishnan @rUv There is no such thing as traditional infrastructure at web scale.” < My point exactly. HackerNewsOnion: Batch is the new realtime. @rbranson : "How do we store this at scale?" "Redis on a cr1.8xlarge" "lmao" John Carmack : We are fundamentally creativity bound at this point. We need faster iterations. ReadWrite : The tech sector as a whole has created more than a tril
6 0.79758388 1117 high scalability-2011-09-16-Stuff The Internet Says On Scalability For September 16, 2011
7 0.79675841 1122 high scalability-2011-09-23-Stuff The Internet Says On Scalability For September 23, 2011
8 0.7883774 1129 high scalability-2011-09-30-Stuff The Internet Says On Scalability For September 30, 2011
9 0.7870338 1484 high scalability-2013-06-28-Stuff The Internet Says On Scalability For June 28, 2013
10 0.77816051 1585 high scalability-2014-01-24-Stuff The Internet Says On Scalability For January 24th, 2014
11 0.77476627 959 high scalability-2010-12-17-Stuff the Internet Says on Scalability For December 17th, 2010
12 0.77423137 1223 high scalability-2012-04-06-Stuff The Internet Says On Scalability For April 6, 2012
13 0.77359837 1134 high scalability-2011-10-28-Stuff The Internet Says On Scalability For October 28, 2011
14 0.77347809 1626 high scalability-2014-04-04-Stuff The Internet Says On Scalability For April 4th, 2014
15 0.77103192 1151 high scalability-2011-12-05-Stuff The Internet Says On Scalability For December 5, 2011
16 0.77058089 1311 high scalability-2012-08-24-Stuff The Internet Says On Scalability For August 24, 2012
17 0.77016366 1190 high scalability-2012-02-10-Stuff The Internet Says On Scalability For February 10, 2012
18 0.76731896 771 high scalability-2010-02-04-Hot Scalability Links for February 4, 2010
19 0.76553011 1649 high scalability-2014-05-16-Stuff The Internet Says On Scalability For May 16th, 2014
20 0.7651782 1203 high scalability-2012-03-02-Stuff The Internet Says On Scalability For March 2, 2012
topicId topicWeight
[(1, 0.165), (2, 0.166), (10, 0.021), (27, 0.165), (30, 0.048), (39, 0.042), (47, 0.012), (56, 0.02), (61, 0.066), (77, 0.041), (79, 0.115), (85, 0.049), (94, 0.023)]
simIndex simValue blogId blogTitle
same-blog 1 0.92559797 835 high scalability-2010-06-03-Hot Scalability Links for June 3, 2010
Introduction: How Big is a Yottabyte? Not so big that the NSA can't hope to store it says CrunchGear : There are a thousand gigabytes in a terabyte, a thousand terabytes in a petabyte, a thousand petabytes in an exabyte, a thousand exabytes in a zettabyte, and a thousand zettabytes in a yottabyte. In other words, a yottabyte is 1,000,000,000,000,000GB . The CMS data aggregation system . The Large Hadron Collider project is using MongoDB as a cache . Here we discuss a new data aggregation system which consumes, indexes and delivers information from different relational and non-relational data sources to answer cross data-service queries and explore meta-data associated with petabytes of experimental data. Google I/O 2010 Videos are up available (many of them anyway). You might be particularly interested in Google Storage for Developers , Building high-throughput data pipelines with Google App Engine , Batch data processing with App Engine , BigQuery and Prediction APIs , and Measure
2 0.92297161 544 high scalability-2009-03-18-QCon London 2009: Upgrading Twitter without service disruptions
Introduction: Evan Weaver from Twitter presented a talk on Twitter software upgrades, titled Improving running components as part of the Systems that never stop track at QCon London 2009 conference last Friday. The talk focused on several upgrades performed since last May, while Twitter was experiencing serious performance problems.
3 0.91224939 705 high scalability-2009-09-16-Paper: A practical scalable distributed B-tree
Introduction: We've seen a lot of NoSQL action lately built around distributed hash tables. Btrees are getting jealous. Btrees, once the king of the database world, want their throne back. Paul Buchheit surfaced a paper: A practical scalable distributed B-tree by Marcos K. Aguilera and Wojciech Golab, that might help spark a revolution. From the Abstract: We propose a new algorithm for a practical, fault tolerant, and scalable B-tree distributed over a set of servers. Our algorithm supports practical features not present in prior work: transactions that allow atomic execution of multiple operations over multiple B-trees, online migration of B-tree nodes between servers, and dynamic addition and removal of servers. Moreover, our algorithm is conceptually simple: we use transactions to manipulate B-tree nodes so that clients need not use complicated concurrency and locking protocols used in prior work. To execute these transactions quickly, we rely on three techniques: (1) We use optimistic
Introduction: This is guest post by Michael DeHaan (@laserllama), a software developer and architect, on Ansible , a simple deployment, model-driven configuration management, and command execution framework. I owe High Scalability a great deal of credit for the idea behind my latest software project. I was reading about how an older tool I helped create, Func, was used at Tumblr , and it kicked some ideas into gear. This article is about what happened from that idea. My observation, which the article reinforced, was that many shops end up using a configuration management tool (Puppet, Chef, cfengine), a separate deployment tool (Capistrano, Fabric) and yet another separate ad-hoc task execution tool (Func, pssh, etc) because one class of tool historically hasn't been good at all three jobs. My other observation (not from the article) was that the whole "infrastructure as code" movement, while revolutionary, and definitely great for many, was probably secretly grating on a good number of
5 0.90279227 1141 high scalability-2011-11-11-Stuff The Internet Says On Scalability For November 11, 2011
Introduction: You got performance in my scalability! You got scalability in my performance! Two great tastes that taste great together: Quotable quotes: @jasoncbooth : Tired of the term #nosql. I would like to coin NRDS (pronounced "nerds"), standing for Non Relational Data Store. @zenfeed : One lesson I learn about scalability, is that it has a LOT to do with simplicity and consistency. Ray Walters : Quad-core chips in mobile phones is nothing but a marketing snow job Flickr: Real-time Updates on the Cheap for Fun and Profit . How Flickr added real-time push feed on the cheap. Events happen all over Flickr, uploads and updates (around 100/s depending on the time of day), all of them inserting tasks. Implemented with Cache, Tasks, & Queues: PubSubHubbub; Async task system Gearman; use async EVERYWHERE; use Redis Lists for queues; cron to consume events off the queue; Cloud Event Processing - Big Data, Low Latency Use Cases at LinkedIn by Colin Clark. It
6 0.89718461 883 high scalability-2010-08-20-Hot Scalability Links For Aug 20, 2010
7 0.89541614 1097 high scalability-2011-08-12-Stuff The Internet Says On Scalability For August 12, 2011
8 0.87469524 900 high scalability-2010-09-11-Google's Colossus Makes Search Real-time by Dumping MapReduce
9 0.87337625 250 high scalability-2008-02-17-Web Accelerators - snake oil or miracle remedy?
10 0.86256272 1483 high scalability-2013-06-27-Paper: XORing Elephants: Novel Erasure Codes for Big Data
11 0.86181802 666 high scalability-2009-07-30-Learn How to Think at Scale
12 0.8562175 555 high scalability-2009-04-04-Performance Anti-Pattern
13 0.85410661 28 high scalability-2007-07-25-Product: NetApp MetroCluster Software
14 0.85014659 1435 high scalability-2013-04-04-Paper: A Web of Things Application Architecture - Integrating the Real-World into the Web
15 0.84061903 1594 high scalability-2014-02-12-Paper: Network Stack Specialization for Performance
16 0.83839607 304 high scalability-2008-04-19-How to build a real-time analytics system?
17 0.83764726 1014 high scalability-2011-03-31-8 Lessons We Can Learn from the MySpace Incident - Balance, Vision, Fearlessness
18 0.83668196 901 high scalability-2010-09-16-How Can the Large Hadron Collider Withstand One Petabyte of Data a Second?
19 0.83652443 576 high scalability-2009-04-21-What CDN would you recommend?
20 0.8356548 842 high scalability-2010-06-16-Hot Scalability Links for June 16, 2010