Lessons learned from evaluating IOTA on IoT devices

Forums IoTStack News (IoTStack) Lessons learned from evaluating IOTA on IoT devices

Tagged: 

  • This topic has 1 voice and 0 replies.
Viewing 0 reply threads
  • Author
    Posts
    • #28690
      TelegramGroup IoTForIndia
      Moderator
      • Topic 2519
      • Replies 0
      • posts 2519
        @iotforindiatggroup

        #News(IoTStack) [ via IoTForIndiaGroup ]


        We found that despite the theoretical scalability of the Tangle, the actual IOTA protocol has relatively high energy consumption. The Proof-of-Work and transaction signing operations are computationally complex relative to the limited capabilities of many IoT devices and and may be impractical on energy-limited / battery-powered devices.

        The results show the time required in seconds (82 seconds on Raspberry Pi ); extrapolating that to energy, we got 54.9 J (joules) for Raspberry Pi, 233.2 J for the Core i7, and 93.5 J for the Nvidia GPU.
        The computational power of the TI LauchPad is 10–100 times less than that of Raspberry Pi, depending on the exact workload. Even if the PoW code could run on the TI LauchPad, it would take an hour or so just to compute the PoW for a single transaction. The devices would also run out of battery in a day or a few weeks at most, depending on the capacity of the battery (100 mAh to 2700 mAh, respectively).
        Raspberry Pi it itself capable of doing some PoW, but it only can do the PoW for 1000 transactions per day, assuming it’s spending 100% CPU resources for the task — so, clearly, it cannot function as a major PoW hub either.


        Read More..

    Viewing 0 reply threads
    • You must be logged in to reply to this topic.