Our Data

Stark Drones wants to practice collecting more intensive data as research and launches progress. Data seen here is "as-is" and mostly been anonymized. As funding and research allows, we want to do more studies for both our algorithms and hardware launches.

Neutrino Oscillations

Finding anomalies from Neutrino Oscillation data sources (CERN/Stanford Experimental Particle Physics Lab), utilizing a novel QuadTree method for zeros between sensitive data points.

Mining Stress Test

Tried finding novel techniques for crypto-mining, offered technical expertise and light algorithmic capabilites to ChainTerra for basic stress testing.

Mining Stress Test

Continued stress testing for ChainTerra on a less congested network.

MSP-430 Memory Configuration

name origin length used unused attr fill---------------------- -------- --------- -------- -------- ---- -------- SFR 00000000 00000010 00000000 00000010 RWIX PERIPHERALS_8BIT 00000010 000000f0 00000000 000000f0 RWIX PERIPHERALS_16BIT 00000100 00000100 00000000 00000100 RWIX RAM 00000200 00000080 00000032 0000004e RWIX INFOD 00001000 00000040 00000000 00000040 RWIX INFOC 00001040 00000040 00000000 00000040 RWIX
See full memory configuration mapping --> at the repository

<?xml version="1.0" encoding="UTF-8" standalone="no"?><configurations XML_version="1.2" id="configurations_0"> <configuration XML_version="1.2" id="configuration_0"> <instance XML_version="1.2" desc="TI MSP430 USB1" href="connections/TIMSP430-USB.xml" id="TI MSP430 USB1" xml="TIMSP430-USB.xml" xmlpath="connections"/> <connection XML_version="1.2" id="TI MSP430 USB1"> <instance XML_version="1.2" href="drivers/msp430_emu.xml" id="drivers" xml="msp430_emu.xml" xmlpath="drivers"/> <platform XML_version="1.2" id="platform_0"> <instance XML_version="1.2" desc="MSP430G2001" href="devices/MSP430G2001.xml" id="MSP430G2001" xml="MSP430G2001.xml" xmlpath="devices"/> </platform> </connection> </configuration></configurations>

github.com/Mentors4EDU/Reinvent-the-Internet

Photon Dev Board Subscription Data

This trace was taken from app data from 6/27/2022 for the same board from the Internet Balloon Launch that was on June 18th, 2022.The data have been scrubbed/anonymized.It gives insight to:

Data/Stream Subscription Capabilities
Stream logging capabilities
Bytesize and Post/Data Capabilities
Logging Capabilities w/ Android and Photon Board
Data Pinging and Sync Capabilities

592000,quic=":443"; ma=2592000; v="46,43"

Cache-Control: private

Content-Encoding: gzip

Content-Type: application/json; charset=UTF-8

Date: Mon, 27 Jun 2022 19:48:31 GMT

ETag: etag-1070197449556-fireperf-fetch-1769032129

You can download the subscription data from the repository and use LogPad to query the events. LogPad can be seen here.
You can also use LogCat or a network visualizer to look at trace history. These bytes were transferred within a short time period. The most complex way to visualize it, is perhaps utilizing a network visualizer toolkit that can do so with nano and millisecond trace precision.

github.com/StarkDrones/ParticleIO-Log

Decentralized Internet SDK Usage Stats

As of November 27th, 2022, the Decentralized-Internet SDK repository has reached over 133k downloads just from NPM alone. On the GitHub Marketplace, it has reached 819 people subscribed to the open-source application (for repo tracking purposes), of which it has an 82% active retention rate based off installed vs. not installed.That SDK is an open-source dev kit and library for building an offline-centric network centered around distributed data sharding. It includes clusterpost and the gridbee-framework, amongst other repos. It also implements BOINC. It is mostly hosted through a single monorepo and is the largest dev kit for distributed and grid computing projects, as well as a decentralized web. According to BOINC's website, a small team wanting to process 100 TeraFLOPS for 1 year, would only pay on average $125k using BOINC and the power of volunteer/grid computing as opposed to $175 Million hosting on Amazon's Elastic Computing Cloud. Since BOINC is open-source, the decentralized-internet SDK aims to take that same philosophy and integrate it into the world of decentralized applications. Stark Drones oversees the distribution of that SDK.By the end of the year 2022, the Decentralized-Internet SDK had over 175k downloads across all installation sources. Stark Drones helps maintain this open-source SDK which is meant to be a competitor to Tor.

Knowledge Preservation on the Blockchain

Outside of its own protocols, and various other places Stark Drones distributes information, it is getting in the habit of using third party services as well like Arweave and the Permaweb, along with Archive.org in order to archive data and research. This makes sure Stark Drones and its related research are preserved for all generations. Some of our repos are also part of GitHub's arctic vault.
Previously, archived via Arweave

Realtime Data Analysis and Load Testing w/ the Cloud

We have ran a test utilizing NextFlow and Seqera on the CHIPseq algorithmic template and custom configurations with a single core max CPU and 6GB max memory in the specs, connected to Google Batch through an API. These are the results. We can potentially analyze 500TB of memory for only $31.38 if this setup is scalable. This is related to cloud/edge computing, and likely an algorithm more tailored towards actual targeting and the data types analyzed, can probably yield even more valuable results.

Network Packets:

Starkcom Portal + Wireshark + ChirpStack

Compression Demos

Image, Video, CAD and Audio Compression

AI Model Training

Testing Setup

Programming Language - LonScript Preview

LonScript Benchmarking

LonScript has been evaluated across 15 benchmark tests, where it outperformed Python by more than 1,434x, C by 23x, C++ by 22x, and Julia by 25x. In some individual tests, LonScript performed exceptionally well, exceeding 6,000x and reaching a peak of 16,781x faster than Python in one benchmark. However, to provide a more accurate representation of overall performance, we present the geometric mean results along with charts for the full set of tests conducted.

These metrics are important because when computing and AI companies rely on outdated programming languages, their speed and overall capability can become constrained.

Google Sites

Report abuse