The Latest from ThreeFold - Cloud

🔗Edge Cloud, not Grid Cloud?

The term Edge Cloud or, to be more precise, Edge Computing is one of the most exciting technology terms of these days.

🔗Edge Computing Definition

According to Wikipedia, “Edge computing is a distributed computing paradigm in which computation is largely or entirely performed on distributed device nodes known as smart devices or edge devices as opposed to primarily taking place in a centralized cloud environment.

In summary, this means that the cloud is divided into smaller pieces and distributed to the edge of the network.

🔗Drivers for Edge Computing

Some latency-sensitive applications require fast response times to work correctly. A good example is self-driving cars. A lot of decisions in autonomous vehicles have to be made instantaneously; the decision to react to the unexpected while driving at high speed cannot afford to be slowed down by data traveling long distances to be processed. To improve the system’s reliability, additional hops, and any further points of failure must be eliminated. Moving to compute as far out as possible to the Edge is the logical choice.

More generically, real-world situations are increasingly relying on an IT component to be involved in some form of decision making where time is of the essence. The current IT setup is that we have highly concentrated IT setups in data centers connected by fast networks to the rest of the world. This setup is rapidly becoming obsolete as we need to have the fastest possible response for everyday activities, which cannot sustain the time delay caused by (extended) network connections. Having actual data processing and storage facilities close to these real-world situations is the next IT capacity evolution.

These thousands of situations are mostly in the IoT area. Still, other killer applications like streaming would greatly benefit from being served from the Edge rather than a distant central location.

🔗Edge Computing Market Facts

Due to the billions of IoT devices driven by various use cases like autonomous driving, a massive demand for Edge Computing is evolving rapidly. According to the Digital Journal, the Edge Computing market is growing extremely fast. With a CAGR of over 35.0% and the market is expected to reach USD 33.75 billion by the end of 2023.
Source: http://www.digitaljournal.com/pr/3394129

Edge Computing and Grid Computing are often used in the same context and sometimes are confused. For this article, we’d like to define the terms going forward as follows:

Grid Computing means a computer network of 100% distributed computer resources like processing power, memory, and data storage of similar or different types to which authorized people can run their computation workloads. That means that Grid Compute nodes are mostly in data centers and maybe dispersed over more extensive regions but are typically still far from the Edge. Compared to Edge Computing, Grid Computing environments are owned by more than one individual or company, and thus, controlled by many.

Edge computing has risen from a need to have data processing and storage as close as possible to real-world situations. The situation could be as simple as recording biometric data by a smartwatch to support the required information for a self-driving car to function as intended. The necessity comes from having local computing and storage capacity - it can be used by multiple use cases (situations) but does not have to.

🔗The Benefits of Edge Computing

Edge computing comes with specific benefits that are mandatory for the next evolution in the IT industry. We have optimized the central model to such a degree that the network connects the central capacity to use cases in the field, the next bottleneck. Edge computing is solving several challenges faced by real-world situations with the following benefits:

🔗Speed

Speed is probably the most crucial benefit of edge computing, as there are plenty of use cases that require near-real-time computing.

🔗Scalability

The scalability of edge computing also makes it incredibly versatile. By partnering with local data centers further out on the Edge, companies can easily target desirable markets without investing in expensive infrastructure expansion. Edge data centers allow them to service end-users efficiently with little physical distance or latency. They also do not constrain companies with a substantial footprint, allowing them to shift to other markets rapidly should economic conditions change.

🔗Availability and Security

Traditional cloud computing architecture is, by definition, centralized, which makes it especially vulnerable to distributed denial of service (DDoS) attacks and power outages. Edge computing distributes processing, storage, and applications across a wide range of devices, nodes and data centers, which makes it difficult for any single disruption to take down the network which is why logically and geographically dispersed systems are often used for DDoS mitigation like in Content Delivery Networks (CDN)

By pushing everything to the cloud, you’re leaving your business open to ISP failures and cloud server downtime. Many mission-critical operations like railroads and chemical plants won’t even use the cloud today. Their server farms are the only way to guarantee uptime.

🔗Process optimisation

Another big benefit is process optimization. If self-driving cars were to use the cloud instead of the Edge, they’d be pushing all the data they gather up to the cloud. But if the Edge makes local decisions, the cloud may not need all that data immediately – or even at all.

🔗Requirements for Edge Computing

When compute and storage capabilities exist on the Edge, we need to overcome several challenges around capacity distribution. Most traditional IT technologies do not allow this as they live in large data centers where there is a lot of “care” for these technologies.

The most significant generic hurdle to overcome is that there is going to be a lot less “care” for the technology to rely on at the Edge. These “devices” have to be self-sufficient to no small degree. If we want to sum up some of the specifics of edge technology, this will come out:

• A typical edge device or node must run an OS that comes in at a low cost or, ideally, for free like some open source offerings. This enhances the proliferation of capacity at the Edge.
• Container technology must be embedded in the OS at the lowest possible level to create the maximum benefit of efficiency from the edge hardware: classic Virtual Machine environments are just too inefficient and resource consuming for edge use cases. There will be Virtual Machine based edge solutions, but these are going to be out there for a certain period, catering for current IT workloads to move to the Edge. The Edge’s correct move can only happen when traditional IT applications split up into pieces and require a lot less capacity to run. Hence, it will convert over time to containerized technology.
• Edge Computing hardware will consist of datacenter grade hardware and all sorts of other types of hardware. The Edge might include some smaller sized data center facilities but also include different types of “homes” for compute and storage equipment. Think of transportation as a specific use case - capacity on maritime vessels, capacity on trains, and vehicles’ capacity. Hardware will have to accommodate different use cases, price levels, form factors, and power consumption.
• Edge hardware needs a lightweight OS that is very close to the hardware components to get the maximum out of compute and storage capacity. Less is more - for real Edge capacity, local people won’t be able to interact with edge capacity - one further step - there will be no need to have remote people access these devices for maintenance purposes. The ultimate edge computing solution will have a significant amount of self-sustainability - where there are no more highly technically skilled people required to keep the capacity available, up and running.
• In most cases, Edge means being remote, and having no access to humans or humans are only too expensive for the specific use case.
• Power outages or no network availability will cause system failures. The ideal OS to deal with this revolves around no people; the zero-touch approach should be stateless and needs to have self-healing capabilities run by automation software robots.
  

🔗Existing Edge Computing Solutions

Today there are a few Edge Computing platforms out there that cover some requirements as outlined above.

• Microsoft, for instance, launched products like Azure IoT Edge and Azure Data Box Edge to extend their Datacenter based Azure cloud offering to the Edge.
• Another attempt to bring centralized cloud technology to the Edge is the Openstack Edge Computing initiative, which is in the concept phase only so far.
• A new OS for the Edge is CoreOs, also known as Container Linux, which runs on nearly any platform, whether physical, virtual, or private/public cloud.
• More recently, there are a few blockchain-driven technologies out there pushing for edge capacity to exist. Examples are:
  • SIA
  • Sonm
  • Filecoin
• Content delivery networks with a lot of “capacity” sitting close to end-users are expanding their capabilities beyond content delivery
  • Limelight
  • Akamai
    
    
    

🔗The Holy Grail

The holy grail would be a self-healing Edge Grid. Edge, because of all the benefits Edge Computing brings to the user: Speed, Scalability, Reliability, Security, Versatility, and Process Optimization. Grid, because it fulfills the distributed workload requirements, a classic centralized cloud would generally be used without a centralized nature and all its disadvantages.

Probably the most critical part of the ideal solution is self-healing. The biggest challenge of an Edge Computing environment is the dependency on local care: humans to maintain hardware and upgrade or fix software components. Self Healing needs to do the job here, which means zero people can be involved.

Only a self-healing mechanism with zero-people involved, completely decentralized, brings down the cost and provides more security to the environment, and makes it highly scalable.

Some of you may think that new blockchain-based technologies may be the holy grail. But they are not.

In most cases, their implementations are too slow, and that’s why it does not make sense to use them for production purposes. Another downside is the complexity of the application caused by stitching together 3rd party technologies. In many cases, these blockchain-based technologies are at a very early stage with beta implementations or even roadmap only type of technology.

The holy grail is something that hasn’t been created before; it needs to be a versatile system that is self-manageable. A large global conglomerate should not own the capacity - it should be owned by many. Therefore, it should have a commercial model where people individually can decide to partake in building this global edge solution and where there is a defined return on investment for those partakers.

A few groups of people, like us, are creating decentralized solutions for this, owned by the many, and creating services for all.

🔗ThreeFold’s technology does not follow the traditional IT model

The Edge internet needs a different approach. So, we started from scratch and shaped a new approach. ThreeFold’s open-source technology addresses the scalability, security, storage, energy, and operating issues by simplifying architectures and not relying on third-party software. This drives down costs and improves efficiencies.

🔗What is ThreeFold’s Technology Toolbox?

ThreeFold’s technology is based on three primary building blocks: Stateless Computing, Self-Healing Technology, and Blockchain technology - this is the core of our “Technology Toolbox“.

• Stateless Computing: a stateless operating system, Zero-OS, enabling distributed hardware to form an Edge Internet grid generating IT storage and compute capacity
• Self-Healing Technology: controls and operates a distributed cloud and reserves, procures, and executes workloads on the grid using our Jumpscale platform. It delivers autoscaling and automation frameworks that enable self-healing and can be adapted for any kind of IT workload.
• Blockchain Technology: Via Rivine - our secure, scalable “proof of stake” blockchain, along with our identity management platform, ItsYou.online. Our blockchain technology enables user authentication, access controls, secure workloads, and heightened security.

🔗Enabling Stateless Computing

The global decentralized internet will be composed of thousands (millions?) of nodes distributed geographically. This means the decentralized internet must be powered by a stateless operating system that can be booted via the Internet (as opposed to living on the node locally). This removes the complexity of upgrades, truck rolls, etc.

Open-source components have been developed, Zero-OS, for the deployment of bare metal machines functioning as Internet nodes and application hosts. Our Zero-OS platform runs containerized applications and software code on secure and distributed IT infrastructure. Zero-OS can be deployed on off-the-shelf hardware by anyone willing to make server and storage capacity available in their markets and within our network.

🔗The benefits of Stateless Computing are substantial, including:

• Performance: lightweight OS without overhead for running bare-metal machines.
• Free: finds the optimal performance of server & storage nodes at no cost.
• Private: no data footprint remains while workloads move from one node to another.
• Distributed: capacity is generated without the need for data centers: At The Edge.
• High-available: easy creation of clusters while applications run in fragmented segments.
• Portability: moving software from one node to another within the grid is instant.
• Secure: open-source distributed software with transparent code. No security backdoors
• Sustainable: energy-efficient OS resulting in very low node power consumption.

🔗Technology that is self-healing & self-driving

ThreeFold’s Jumpscale framework and ZeroBot allow the creation of blueprints to auto-deploy applications within the compute environment. The bots are constantly checking the health and uptime of the infrastructure, and runtime environments, to guarantee the performance and availability of applications at all times.

🔗The benefits of our self-healing technology protocols include the following:

• Self-Healing: the platform regenerates when health issues that cause downtime for infrastructure and applications are detected.
• Intelligence: agents capture information from the grid and make intelligent decisions to provide an ideal and optimal host environment for software code and containerized applications.
• Lifecycle Management: the status of the infrastructure and the lifecycle of running applications are constantly checked and recorded.
• Autoscaling: the platform automatically scales to meet application and infrastructure performance.

🔗Blockchain Technology to execute workloads and applications via smart contracts

The engineers developed new blockchain algorithms based on proof of stake for validation purposes and turned this into our Rivine open-source blockchain software.

Unlike the “proof of work” blockchains that currently dominate the space, our Rivine blockchain is scalable, green, and secure - and does not rely on huge mining resources. We run our blockchain on approximately 200 dedicated nodes and do not rely on any third-party software.

In addition to our blockchain technology, we have also developed ItsYou.Online, an identity, and access management platform. Itsyou. online uses three-factor authentication to manage access on our network and facilitate execution.

🔗The benefits of TF blockchain technology and identity management platform include:

• Sustainable: our algorithms to reach consensus use a minimal amount of energy and electricity compared to other blockchain software and networks.
• Lowest possible cost: of transactions make it an efficient and fast payment system.
• Neutral: state of the network and applications are recorded in a transparent and immutable manner.
• Payment fulfillment: transactions are automatically paid using ThreeFold Tokens.
• Smart Contracts: fully automated execution of contracts from customers buying Internet IT capacity.
  
  All of our technology is open-source and can be accessed in the following Github Repos:
  
  

• Zero-OS: https://github.com/threefoldtech/zos
• Jumpscale: https://github.com/threefoldtech/js-ng
• BCDB : https://github.com/threefoldtech/bcdb

🔗Hyperscale realities are very different from what most people think

🔗Myth 1: Bigger data centers are more efficient

There are advantages in building larger-scale data centers, like economies of scale and sheer bulk buying power, but these are not as significant as what people think. The average cost per rack in a hyper-scale data center is $20-35K USD including all energy requirements and safety systems. The cost of hardware per rack is around the $200-300K USD mark.

What is often forgotten is that anything that is of enormous scale and highly concentrated comes with complexity and specific problems to deal with. For example. resource requirements such as investment, operational costs, knowledge, and people increase significantly with size.

In reality, keeping things simple, small, and distributed is much more cost-effective than large complex environments.

🔗Myth 2: Big data centers can be sustainable

The carbon footprint of a big data center is enormous. To improve the power usage effectiveness (PUE) of most data center farmers have adopted wind, hydro, and/or solar power technologies, which indeed helps drop their PUE by an estimated 20%. But is this leading to more sustainability?

This 20% looks great on paper and in the farmers’ corporate social responsibility reports. However, it just represents an improvement in the cooling technology and sourcing of energy. It doesn’t actually impact the energy consumed by the equipment that runs in their data centers (servers, storage chassis, physical disks, etc.) which is the equipment responsible for the carbon footprint. PUE only refers to overhead power consumption, i.e. cooling the facility, opening, and closing doors, maintaining power security systems, etc.

Real improvement lies in deploying technologies that actually consume less power to deliver the actual Internet capacity to run workloads, real CPU chassis, physical disks, and storage cabinets. Improving on how hardware is being more effectively used can have an impact of up to a thousand percent and lead to ten times more power-efficiency.

🔗Myth 3: Redundant systems have better uptime

A lot of us believe that systems need redundancy mechanisms to improve their operational uptime and reliability. While it may make sense to IT experts, let’s translate this for the non-IT world?

To make a car more reliable we add redundancy (as we do in IT). So for the risk of having a puncture, we add one extra tire for all the tires we use continuously. This adds 4 extra tires to the car. Then a decision needs to be made: Do we put those tires in a structure where they are always running along with the primary tires or do we choose not to have them “online” all the time, wearing and tearing in the same way as the primary tires?

Building such a system would take a large number of engineers to come up with a solution, and would change cars as we know them. Wouldn’t it make more sense to think outside the box and solve the root of the problem by making tires un-deflatable?

The IT industry has gone overboard with the concept of redundancy, having forgotten to look at the root cause issues. This has spawned a whole new industry of itself, which has a financial interest in creating complicated and expensive redundant systems.

🔗Myth 4: Big companies optimize better

Big companies with a certain track record will know better how to optimize as they have more people and resources.

At first glance, this sounds logical, but in reality, ninety percent of the innovation within the IT space comes from startups. The giant IT companies have a heritage they hardly can overcome. They are locked in old infrastructure designs, and building outside of that infrastructure would be costly, timely, and probably put the breaks on their businesses.

Real innovation gives way to fix the core symptoms rather than taking the problem pain-killer approach.

🔗Respect Human Rights

• Sustainability: It is important for a very simple reason: we cannot maintain our quality of life as human beings, the diversity of life on our planet, or Earth’s ecosystems unless we embrace and protect its balance.
• Equality: Everyone should have a right to learn, dream, partake and succeed. Equal access also means affordability, not ownership by big greedy companies.
• Decentralization: To better distribute power, value and opportunities, we need more decentralization. Technologies and solutions should be available everywhere to improve equality and sustainability.
• Openness & Interoperability: Allows technologies to work together in an open way, without limitation, and therefore expands the possibilities for innovation and sharing.
• Accessibility: Whatever the location or situation, people should benefit from easy and fast access to information and solutions.
• Security & Privacy: Protect people from being put in situations of danger or exploitation. People should be empowered to be independent and they should be the sole owners of their experiences.

🔗Respect Human Effort

• Reliability: Make sure technologies are reliable and last longer.
• Simplicity: We need to keep things simple: no unnecessary complexity - less is more.
• Transparency: Let people and companies work together in a 100% transparent way.
• Collaboration: It’s all about the greater picture. We are stronger together and have to help each other in reaching our common visions.

🔗Respect Human Experience

• Ergonomy: Products and solutions should empower productivity, safety and comfort, while reducing human error.
• Inclusiveness: Design should always be global and inclusive. We are all equal.
• Design: Products should be visually appealing and should achieve it’s purpose.
• Transparency: Organisational records and strategies should be accessible to everyone.
  
  Photo by Renee Fisher on Unsplash.

🔗An introduction to the problems

• The Internet is no longer the neutral place it used to be. +80% of the Internet capacity is delivered by a handful of companies. This results in a loss of neutrality and openness and breeds inequalities.
• To support the continuous growth of the Internet and connect billions to new Internet Of Things devices, 400 million new servers are needed by 2020. The world therefore needs another 4,000 large-scale data centers that each consume 200 MWe of electricity: see source.
• Today, about 4 billion people have no access to the Internet due to the high costs of connectivity and capacity. We believe these people have the fundamental right to have low-cost access to digital resources such as online identity solutions, education, health information and more. For this to happen, solutions that can lower the cost by a factor of 10 are required.
• Bandwidth does not follow Moore’s law. The growth of bandwidth required is much higher than the speed at which new capacity can be delivered. This is even more important in emerging markets.
• Current blockchain technology is not scalable and not sustainable. Example: the Bitcoin blockchain alone uses more energy today than Croatia.

🔗Current Internet and Cloud capacity

Today, the cloud computing industry represents a 100 billion dollar industry - mostly owned by about 20 global cooperations. Experts such as Bell Labs and Cisco believe the capacity requirements in the next decade will dramatically exceed existing infrastructure - requiring many more data centers and billions, if not trillions in investments.

The spectacular growth and advancements in cloud computing that occurred over the last decade resulted in large part from “hyper-scale” efficiencies - i.e. centralizing Internet infrastructure in very large billion dollar data centers. However, demand for Internet capacity already surpassed the available infrastructure with emerging technologies such as Big Data, the Internet of Things (IoT) and Artificial Intelligence (AI) Technology. The existing hyperscale cloud architecture cannot cope with this growing demand as the requirements for land, power and funding are becoming outrageous.


The challenges facing the Internet includes the following:

• Trillions of Investment Required. In the next decade, the global Internet infrastructure requires another 4,000 data centers (250M - 1B dollar investment each) to meet demand which would cost trillions - this money could be better used in education, healthcare, or sustainable innovations.
• Impossible to scale the networks connecting users to the hyperscale data centers. Since hyperscale data centers are built in large “central” locations, the network leading to these data centers is congested. Even more, remote regions that are less connected to the Internet cannot benefit from these data centers’ IT services.
• Non-local data and capacity. Governments around the world are becoming more and more concerned about data leaving their borders. To tackle this centralization issue, they would need local IT infrastructure. Also, IoT workloads need local capacity to deliver constant real-time computing and storage requirements - for certain applications and workloads (e.g. driverless cars), latency is not an option. Hyperscale data centers can simply not match these requirements.
• Not secure / at risk of choke point failures. There are plenty of recent well-publicized problems around security, outages and loss of data. The reason for this is that centralized hyperscale architectures are inherently insecure - and failures at a single hyperscale location can cause widespread contagion and outages.

🔗The ThreeFold Solution

Our solution to these trillion dollar problems is to migrate Internet capacity to the Edge (closer to user location) - i.e. a “hyper-distributed” network architecture, located close to users (in cities, schools, offices, homes and even vehicles).

Edge networks will not necessarily replace the existing hyperscale capacity powering today’s cloud - rather, they will scale to meet the increased capacity demands going forward. However, if the current hyperscale capacity can be decentralized to improve energy consumption and reduce costs, this would be a huge global success.

By 2025, 60% of new Internet infrastructure will be decentralized and located at the edge of operator networks by using alternative locations which have stable (renewable) power and reliable connectivity.

🔗ThreeFold’s technology main benefits:

• Security: The ThreeFold Grid is distributed and benefits from our ItsYou.Online identity management platform - making it much more secure.
• Decentralized: The solution enables to bring capacity closer to where users are located, enabling a Cloud at the Edge.
• Sustainability: The technology removes the need to build huge power hungry datacenters and by decentralizing the architecture enables up-to 1000% energy savings.
• Making IT a true utility: We always approached IT as Energy. Today we provide 3x more uptime and between 3-10x more cost effectiveness.
• Neutrality: The solution enables 100% neutral & private Internet IT capacity. Our blockchain technology provides security, neutrality & scale.

🔗The ThreeFold Zero-Node

The Zero-Node provides compute, network and storage capacity to the internet and ThreeFold’s blockchain. This capacity can be used for any possible workload. Here are some characteristics of ThreeFold’s Zero node:

• Modularity: Scaling from 8 vCPUs and 3.5 TB of storage (smallest) to clusters with 8000 vCPUs and 2 PB of storage (large), there is a range of possibilities at hand.
• Sustainability: Storing the hardware uses about 10x less energy, and about 2-3x for other workloads.
• Decentralized: The Zero Operating System (“Zero OS”) is ideally suited for Edge networking - where automation, self-healing and hyper-distribution are must-haves.
  • The Zero-Nodes are deployed as close as possible to the consumers of Internet IT capacity (IoT, residential, business, government, education, …).
  • This results in massive bandwidth savings, we estimate to be able to keep 99% of the required bandwidth locally which has a huge impact on cost.
  • This is the only viable solution to be able to provide internet to the 3.2 billion people waiting.
• Self healing: The architecture is designed to ensure that workloads and data can be accessed at all times.
• Neutral & private: A lot of work has done using blockchain (technology which powers crypto currency) to guarantee privacy and net neutrality.
  • Designed around simplicity to be used by everyone.
  • The technology provides anyone with the autonomy to deliver complex IT applications without the need for complex integration services.
• Empowers equality: Our technology should be usable by the “other billions” that remain unconnected. Due to the low cost-points, low energy requirements and reduced need of backbone network capacity.