HashCore Pulse 1.0.0-alpha1
When a fleet of a thousand ASICs is scattered across multiple locations, the most common problem is not a lack of data, but too much of it without context. Everyone has metrics: hash rate, temperature, consumption—any tool that can query a miner's API can read that. The difference begins where the table of numbers ends: in whether the system distinguishes scheduled maintenance from failures, allows for safe simultaneous restarts of five thousand devices, and speaks the language of the operator, not just the exchange.
HashCore Pulse is a cloud platform for monitoring and managing industrial ASIC fleets, designed to scale from a single rack to hundreds of thousands of devices across multiple sites. In July 2026, we released the first version—1.0.0-alpha1—and here we explain what went into it and why this is only the beginning.
An ecosystem, not just a standalone product
Pulse is not our first development. HashCore has been creating its own firmware for ASIC devices for several years (HashCore Firmware), the HashCore Toolkit desktop application, and the free HashCore Proxy Stratum proxy. Pulse is the fourth product in the ecosystem, and in the future, some of the Proxy functionality will be integrated directly into the platform itself.
Today, Pulse manages Bitmain and MicroBT (WhatsMiner) hardware running Stock and HashCore firmware. Separately, without active control, the platform also displays statistics from devices running Braiins, Luxor, and ePIC firmware—meaning that even a fleet with mixed firmware is visible in one place, even if not all of it can be managed yet.
Devices are discovered automatically—no manual addition needed
Each site (farm location) is configured separately: its own electricity tariff, power limit, and display currency for financial indicators. At the same time, equipment is not added manually—the agent on the farm server scans the specified IP ranges, discovers the devices itself, and then simply collects metrics from them.
If the farm previously ran on a different system, migration does not require rebuilding the equipment list from scratch: Pulse can import network ranges from HashCore Toolkit, Hive OS, Awesome Miner, and Bitcap. And if the scanning task (ScanTask) is later deleted, all associated devices are completely removed from the system—rather than hanging in an Offline status.
Offline ≠ failure
In most monitoring systems, device status is a single field. In Pulse, there are two, and they are independent of each other.
Connectivity is solely responsible for whether there is communication with the device. Miner Status is responsible for what it is physically doing: mining, stopped, undergoing autotuning, or in error. If power was lost on a rack at night, that is indeed Offline, and an alert should trigger. But if a miner is simply paused while remaining healthy and online, that is a different state, and a redundant alert is not needed.
It sounds like a small detail, but it is precisely because of the confusion of these two concepts in competing systems that real failures often drown in dozens of planned stops per shift.
Maintenance is not an incident
An extension of the same concept is the Service mode. When a technician removes a device from a rack for repair, it is set to a separate status specifying the reason for the downtime. While the device is in service, it is excluded from error statistics and automatically skipped during group operations—with a clear notification of how many devices were excluded and why.
A separate history is saved for each service session: when it started, what was done, how long it took, and who was responsible. Both a single device and an entire rack can be put into this mode in a single operation. For operators managing client equipment, this is especially valuable: the service log becomes a transparent document rather than a set of red dots in a general log of failures.
See what is needed now, not just a table
A list of thousands of devices is useless if you have to scroll through cards one by one for every check. In Pulse, the table can be instantly switched to a mode displaying the configuration of each device's pools: which pool is currently active and how the others are configured. There is no need to open each device separately to find those with drifted pool settings.
Columns for sorting are grouped logically—hash rate, temperatures, network, firmware—instead of a flat list of twenty unrelated fields. The event log for each device records changes in a "before → after" format: if someone changed the overclocking preset a week ago, the exact values before and after are visible, rather than an abstract "settings changed" entry.
A specific feature for those physically working with racks: the IP Report button on the ASIC device itself allows technical staff to immediately identify a specific machine in the interface—press on the chassis, and the device appears in a separate list in Pulse. It also works in reverse: the Locate command helps find the required device physically among hundreds of similar ones on a rack.
Bulk operations that don't crash the grid
Reboot, change pools, update firmware, manage hash boards and cooling—any operation can be applied to either a single device or an arbitrary group. However, a command to five thousand devices is not sent in a single burst: Pulse staggers it in batches with a configurable delay between them, so that a simultaneous restart of the entire fleet does not create a peak load on the site's power grid.
Once the operation is complete, the exact success percentage is shown, and if some devices failed to execute the command, they can be isolated with a single click without reviewing the entire list manually. Group management of hash boards and cooling additionally supports conditional logic—for example, disabling a specific board if it is in an error state or a temperature threshold is exceeded—and partial updates, where empty fields do not affect the device's current settings.
Comparison with rated performance
In addition to the current and average daily hash rate, the device card displays the factory (stock) rate—a reference point for comparison. If a batch of equipment falls short of the declared characteristics, it is not a vague feeling of "seems to run hotter," but a specific percentage of the rated power that can be presented to the supplier.
Finance in the operator's language
Site profitability is calculated not only in dollars but also in the mined coin—for many operators, this is more important than the exchange rate on a specific day. The Bitcoin and Litecoin exchange rates are visible right on the dashboard, without switching to an exchange website.
For operators outside the US, site financial indicators—revenue, electricity costs, analytics—can be fully translated into rubles, euros, yuan, or tenge. Among Western competitors strictly tied to the dollar model, this is a rarity.
Account protected at a level familiar to enterprise
Two-factor authentication, a list of active sessions with the ability to terminate any of them remotely, password reset, and account deletion in accordance with GDPR—a basic set of features that in this segment often turns out to be missing rather than basic.
Where we are now
Pulse is in alpha—we are launching the platform early on purpose to build it together with those who will use it. New releases come out regularly, and what goes into the next one is largely determined by feedback from real operators.
Currently in development: pool and filter presets, group management of device presets, scheduled automation and conditional triggers, exporting data from the device table, user grouping of devices, technical issue tracker, roles and access permissions, reports, email and Telegram notifications, and embedding the agent directly into the HashCore firmware so that device discovery works out of the box without a separate installation.
Connect, test the features on your farm, and send feedback—it directly affects what goes into the next release.
Registration: https://pulse.hashcore.com/auth/register
