Can Cloud Computing Actually Save the Internet?

Techwriter10 0 Tallied Votes 534 Views Share

Cloud computing or software as a service (SaaS) has a lot of advantages, especially for small and medium sized companies with fewer resources to devote to an IT infrastructure. The cloud vendors, whether , Amazon, Salesforce.com (or whatever company) provide all the software and back-end infrastructure support. They even update the software on the fly sparing your company many of the headaches associated with implementation and ongoing maintenance.

Given that all the transactions are actually crossing the internet, however, it would make sense that it would simply add to the increasingly clogged byways of the internet. But when I spoke to representatives from Google and Salesforce.com at the Enterprise 2.0 Conference in Boston last week and asked them about this (in separate conversations, I might add), I was surprised to hear them argue the opposite—that Cloud computing could actually *reduce* traffic. Confused? Read on.

As a bit of background, the idea of a clogged internet has been on my mind recently. At the beginning of May I published Will P2P Really Kill the Internet? and a week later I published What Does the Future Hold for the OS? in which I argued that the OS will become increasingly less important as we do more of our computing in the cloud, that is, on the internet. In that post, I got a comment that stated “It won’t happen. I doubt the net could handle that much traffic.” Well, just to show that I pay attention to what my readers say, that thought stuck with me. And last week when I was at an Enterprise 2.0 conference event where representatives from Google, Amazon and Salesforce.com pitched the idea that a company could move *all* of its IT infrastructure to the cloud to a group of skeptical CXOs, the clogged internet argument popped into my head.

The CXOs brought up a lot of good arguments as to why they could not move certain information to the cloud, but they never asked about internet traffic, so after the event was over, I walked up to the stage and approached Ross Piper, SVP of Enterprise Strategy at Salesforce.com and later Jeff Keltner, Business Development Manager at Google Apps. I asked them what would happen if they actually succeeded at attaining this vision and most applications moved onto the internet. How would this affect the increasingly clogged pipes of the internet?

Both argued in separate conversations that it would actually reduce traffic because instead of moving large files around many times, you are actually moving around references to the files sitting on their back-end servers and most of the heavy lifting would not be on the internet itself, but on the company server infrastructure. What’s more, they also argued what I call the Jay Leno – Doritos argument. You may recall back in the 90s, Leno made an ad for Doritos in which he said, “Don’t worry, we’ll make more.” When it comes to internet capacity you could make a similar argument, as demand increases the infrastructure will adjust and we will make more.

If you buy this argument, you could see where increasing use of the cloud actually reduces the pressure on the internet pipes as people stop moving large documents around using email and instead point to a file on cloud vendor’s servers. It might just work, and if it doesn’t, we can at least hope for the Leno effect. We’ll make more.

Dani AI

Generated

Good framing from — cloud delivery changes where data moves, but it is not a magic fix for capacity problems. Centralizing files and using shared links can cut duplicate transfers, yet cloud use also introduces new traffic patterns (replication, backups, telemetry, chatty sync clients). The net effect depends on architecture and operational choices, not on “cloud” alone.

Technically, the things that actually reduce Internet load are: single canonical storage (one copy served to many users), server-side processing instead of client downloads, aggressive edge caching/CDNs, delta or chunked sync (send only changed bytes), streaming rather than full downloads, and efficient protocols with proper cache headers. Conversely, uncontrolled client polling, multi-region replication without planning, and naive backup/sync schedules create storms. When evaluating a move, measure bytes-per-user today and after a narrowly scoped pilot before wider rollout.

Practical troubleshooting and checklist: 1) Baseline network usage (NetFlow/SMNP/usage graphs). 2) Pilot one team, track top talkers and sync behavior. 3) Enable caching and set correct ETag/Cache-Control headers. 4) Use delta-sync or chunking for large files and throttle concurrent syncs. 5) Schedule bulk transfers off-peak and limit auto-backups. Watch for clients that continually retry or reupload identical files — that is often where traffic balloons.

For tech-support firms like the question from , the opportunity is to move up the stack: offer migration and hybrid-cloud design, WAN optimization, monitoring/SLA management, security/compliance packaging, and on-site caching appliances or direct-connect setups. Ignore off-topic promos such as ’s marketing post; focus on measurable reductions and service offerings that solve real network pain points.

kladizkov 0 Newbie Poster

how can a technical support company survive the cloud computing...

Andy Zhao 0 Newbie Poster

How big China SaaS market is? Capture a Seat with PC Stars partnership!

It's Andy Zhao at PC Stars, which is a China leading provider of distribution and Hosting services for SaaS and on-premises ISVs.

Having been in this field for years, we believe and witness that there are huge needs for software in China. Unfortunately, piracy has always been an obstacle for on-premises vendors. Since SaaS delivery is getting more and more popular, it’s not a problem anymore. We are convinced that it’s the right time to enter China market for SaaS vendors.

If China is on your recent roadmap, we invite you to join us, exploring Chinese marketing together.

As the sole operation provider for ZOHO ( ), we’re providing a comprehensive hosting and distribution service for ZOHO. We are also helping other SaaS or on-premises ISVs getting success in China market, such as Siondo, Kaspersky, Intel Software, TechSmith, QSR and so on. With the experience accumulated in years, we feature a Demonstration Farm Model of Software Publishing through our service suites as followings:
1.)Localization.
We have a team dedicated to localization, covering apps, websites, documentation etc.
2.)Marketing campaigns
Including but not limited in:Buyers Targeting and Buyers Survey; Sales Lead Generation & Potential Acquisition Solution; Prospect Maturation Solution
3.)Local Deployment & Hosting
For SaaS Vendors only.

It would be great to have the opportunity to discuss the huge opportunities in China Market.
If you have any question, please do not hesitated to contact me.

Many thanks,
Andy Zhao
PC Stars
E-mail:
Phone: 0086 10 59937000 x. 8241
Website:

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.