We’ve all heard the Apple Slogan “Think Different” but is that something we actually do? In this Network Collective Short Take, Russ and Eyvonne explore this idea of thinking differently and how our thought process impacts the work that we do.
The post Short Take – Thinking Differently appeared first on Network Collective.
Who supplies your Internet? If you live in urban Africa, you probably get Internet access through your mobile phone or through fibre at the office or home. When you travel to rural or underserved areas, there is probably limited or no Internet because mobile network operators and Internet Service Providers (ISPs) have yet to reach these areas. But what if people in rural and underserved areas could build their own networks to provide the telecommunications services that they need?
This notion has been tried successfully in several African countries. In Kenya, Tunapanda Institute has built TunapandaNET a wireless network connecting schools and youth centres in Kibera, Kenya’s largest slum. Bosco is a solar-powered wireless network that connects community ICT centres in Gulu, Northern Uganda. PamojaNet operates a wireless network to the community in Idjwi Island on lake Kivu, Eastern DRC, close to the border with Rwanda. Others such as Machaworks in rural Zambia and Zenzeleni Networks in Eastern Cape of South Africa provide similar services. These networks have been built by local communities to provide access to both offline content and Internet access where possible to the communities that they serve.
During a session at the 2018 Africa Internet Governance Continue reading
Cloudflare has a cloud computing platform called Workers. Unlike essentially every other cloud computing platform I know of, it doesn’t use containers or virtual machines. We believe that is the future of Serverless and cloud computing in general, and I’ll try to convince you why.
Two years ago we had a problem. We were limited in how many features and options we could build in-house, we needed a way for customers to be able to build for themselves. We set out to find a way to let people write code on our servers deployed around the world (we had a little over a hundred data centers then, 155 as of this writing). Our system needed to run untrusted code securely, with low overhead. We sit in front of ten million sites and process millions and millions of requests per second, it also had to run very very quickly.
The Lua we had used previously didn’t run in a sandbox; customers couldn’t write their own code without our supervision. Traditional virtualization and container technologies like Kubernetes would have been exceptionally expensive for everyone involved. Running thousands of Kubernetes pods in a single location would be resource intensive, doing it in Continue reading
Welcome to Technology Short Take #106! It’s been quite a while (over a month) since the last Tech Short Take, as this one kept getting pushed back. Sorry about that, folks! Hopefully I’ve still managed to find useful and helpful links to include below. Enjoy!
In this video, Tony Fortunato tests WinPcap versus Npcap. See which performs better.
After explaining the basics of SD-WAN, Pradosh Mohapatra, the author of SD-WAN Overview webinar focused on SDWAN reference network design.
You need at least free ipSpace.net subscription to watch the video.
ApproxJoin: approximate distributed joins Le Quoc et al., SoCC’18
GitHub: https://ApproxJoin.github.io
The join is a fundamental data processing operation and has been heavily optimised in relational databases. When you’re working with large volumes of unstructured data though, say with a data processing framework such as Flink or Spark, joins become distributed and much more expensive. One of the reasons for this is the amount of data that needs to be moved over the network. In many use cases, approximate results would be acceptable, and as we’ve seen before, likely much faster and cheaper to compute. Approximate computing with joins is tricky though: if you sample datasets before the join you reduce data movement, but also sacrifice up to an order of magnitude in accuracy; if you sample results after the join you don’t save on any data movement and the process is slow.
This paper introduces an approximate distributed join technique, ApproxJoin, which is able to sample before data shuffling without loss of end result accuracy. Compared to unmodified Spark joins with the same sampling ratio it achieves a speedup of 9x while reducing the shuffled data volume by 82x.
The following charts show ApproxJoin’s latency Continue reading
Broadcom closed on its $18.9 billion acquisition of CA earlier this week.
The 5G-related network infrastructure market will grow from about $528 million in 2018 to $26 billion in 2022 at a compound annual growth rate of 118 percent, predicts IDC.
While Cisco adds Kubernetes support across its portfolio, other vendors like VMware and IBM recently bought companies to boost their container cred.
The platform minimizes the need for cloud resources to power edge devices by turning ordinary devices, like a mobile phone, into a mini cloud server.