Hypernet is a protocol for high performance computing across a distributed network of devices. Hypernet provides an API suite built on the principle of distributed average consensus (DAC) and tailored for large-scale distributed compute environments. Compute power sellers can earn by selling unused computing power from the devices they already own.
Distributed.fund for PALMINA Invest
For now, the market of cloud computing is dominated by large centralized entities such as Microsoft and Amazon. Hypernet proposes cost-efficient solution excluding hardware expenses, infrastructure maintenance, costs for staff to maintain, manage and monitor infrastructure and the same time making possible to leverage unused computing power across personal devices to handle the continuously growing amount of data transfers.
The majority of Hypernet’s competitors implement a linear approach to computing. DAC, which powers the Hypernet, takes a new approach that has been created specifically to perform parallel computing. Popular use cases could be machine learning, sequencing, data processing and more.
Key competitors:
Existing models are not well constructed to solve general computing problems on distributed networks. The known state and availability of each machine are unknown and changing in time. This issue forces projects as Golem to focus on specific tasks to reduce another layer of overhead.
DAC + Blockchain allows the efficient distribution compute jobs, and effectively manages computers dropping on and off the network. Founders of Hypernet believe that many computers do not use the full capacity. Buyers will have computing requirements where the seller must prove that he capable of meeting the buyer’s needs. Otherwise, the ambitious goal of achieving broader computing is hard to finalize in practice.
Hypertoken will serve as a mechanism for buyers to pay sellers for their computing time providing for users possibilities to stake network peers and track their reputation in the ecosystem.
Key functions:
Collateral - Buyers and sellers of computational power will use tokens as a form of collateral for launching and fulfilling compute jobs. This will ensure fairness for all parties and prevent bad acting.
Capacity measurement - Hypertokens will be used to allocate and measure capacity for submitting and running compute jobs.
Reputation - This will encourage ecosystem actors to behave in the appropriate manner and build proven verifiable reputation across the network.
Governance - Participating nodes are incentivized for helping maintain the quality of the network. Major changes in the network may be voted on.
Daniel Maren (Co-Founder) - Daniel studied computer science at Stanford University. He founded a solar power electronics company, Dragonfly Systems, which was acquired by SunPower Corporation in 2014. Daniel remains an advisor to SunPower, guiding solar and storage product efforts and international business development. An expert in international infrastructure development, finance, and energy, he is a Forbes 30 under 30 award recipient.
Ivan Ravlich (Co-Founder) - Ivan pursued his masters in aeronautics and astronautics at Stanford, where he designed, built, and tested hybrid rockets. Ivan has worked for several startups, notably AdAstra, a plasma rocket startup and LanzaTech, a alternative fuel startup. His PhD research at Stanford focused on advanced space propulsion from magnetoplasma rockets to extended theories of gravity.
Todd Chapman (Co-Founder) - Todd Chapman was awarded a Department of Defense National Defense Science and Engineering Graduate fellowship in Stanford’s Aeronautics and Astronautics department, where his PhD thesis centered on nonlinear model order reduction methods.
Christopher Hansen (Lead Engineer) - Christopher Hansen studied electrical engineering and computer science at Stanford University. After completing his Master’s degree, he joined NVIDIA to work on next generation GPU technology. Drawing upon this experience, in addition to previous work at Facebook and Twitter, Christopher works on the cutting edge of the infrastructure that powers machine learning and HPC.
Randall Kaplan – Advisor / Founder and CEO at Sandee
Randall is a co-founder of Akamai Technologies, providing Content Delivery Network services, founder and CEO of JUMP Investors, a venture capital firm that invested in more than 60 early-stage technology companies.
Tony Reeves – Advisor, CFO at Experian Global Technology
More than 20 years of experience as a CFO
Joseph Urgo – Advisor, Co-founder at district0x
Joe previously spent three years as an Operations Manager for Coinbase
He also advised projects in crypto space: Status, Keep, Bloom, Witnet, Access Network, Nebula, LiquidX, Fabric Ventures, and Republic
Team revealed private demo of their product, everything worked fine, no issues or vulnerabilities were found. While conference call team members gave all answers on specific questions of our tech specialists in confident and professional way.
Mainnet launch is going to be in the middle of the August, 2018.
Telegram – 28 000 members
Twitter – 1 900 followers
Medium – 317 followers
Reddit – 597 subscribers
Hypernet got reviews from some crypto researchers: Pick A Crypto, Coin Crunch, and others. Project evaluations can be found in some popular crypto spreadsheets. They have respectful investors onboard, such as Torque VC, Aritisian VC, but others are undisclosed. There is no oversubscribing interest for now but community is growing. Considered hype level is Medium.
Demand for computing power is continuously growing and Hypernet fits the overall industry trend. Role of the Hypertokens makes sense and adds real utility for the whole ecosystem. Project concept has own advantages and progressive approach compared to existing analogs. New projects in this area appear on the horizon and long-term results will also depend on the possibility of team members to maintain network and deal with issues over a long period of time, there is no code available for open source for now so users should trust the development team and rely on their skills. Hypernet team consists of professionals with relevant experience in computer science. Token metrics are attractive, not very high Hard Cap and Valuation, balanced circulating supply. Managers are open for conversation and actively interact with their audience, presenting video content, key features of Hypernet and text reports. Future performance will strongly depend on the possibility to attract large consumers to participate in the network. There are a lot of users who can provide computing resources, but their consumption depends on entities who are ready to use it. There is no clear information about such interest for now. In case of lack of consumers and according to overall market conditions Hypernet may be highly speculative in the beginning.