# Sprints: Intermittent Blockchain PoW Mining

## Special Note for this blog
This blog is written by group of 4 pepople
1. Shreekar (B23CS1069)
2. Tavishi (B23CS1101)
3. Jyothsna (B23CS1076)
4. Hema (B23CS1031)

So if some text is written by someone, then they will first write their name at top of thier written text in parentheses.
example :

*(person_name)*
some text by person_name

## Blog

### The problem statement
*(Shreekar)*
today our group started breaking down the "sprints: intermittent blockchain pow mining" paper from the usenix security 2024 symposium. the main problem this paper is trying to address is the massive environmental crisis caused by legacy proof of work (PoW) cryptocurrencies like bitcoin. right now, these networks are incredibly successful at securing billions of dollars in digital assets without needing any central bank or government to oversee them. they achieve this through the nakamoto consensus. but this objective security comes at a really crazy cost to our planet. the paper highlights a wild statistic: the global bitcoin network consumes more electricity than the entire country of argentina just to keep the network secure and process transactions. because of this huge energy usage and massive carbon footprint from both manufacturing and running the hardware, lawmakers and regulators around the world are getting seriously worried. some countries are even considering putting strict bans on mining operations altogether. so there is a massive pressure on the blockchain community to find a greener way to do things, but without losing the mathematical security that makes pure PoW so valuable in the first place.

*(Tavishi)*
On thinking more deeply as to why this problem is so hard to solve, we find that a lot of people think we can just fix the environmental issue by tweaking the network settings, like changing the block size or making blocks much faster. However, the paper points out that this actually does not work at all. The amount of energy used is tied to economics and not code parameters.
Rational miners will always use as much physical power as they possibly can, as long as the block rewards they get are profitable. If the puzzle is made easier, miners can simply buy more machines to solve it faster.
Thus, the researchers realized they needed to find a brand new protocol design. They wanted to keep the exact same security guarantees of pure PoW (where anyone can join and verify the chain using just math) but drastically cut down the physical energy consumed by the network. Their ultimate goal with this paper was to achieve an order of magnitude lower carbon footprint while keeping the system totally decentralized and open for anyone to join permissionlessly.

Sprint : We assume all miners have similar, efficient, PoD hardware; others with weaker hardware would not participate.

### Some early understandings/learnings:

*(Tavishi)*

#### VDFs at a glance:

> **VDFs => Verifiable Delay Functions**

**Input** : A complex, sequential problem which takes a predictable amount of time to compute.

**Output**: There is only one output for a particular input. Whatever is the output, based on that a winner is chosen.

**Advantage**: Even if a hacker finds out what the input was, and starts computing the final result (and accordingly either changes his own properties in order to be declared as winner, OR, manipulates the input in his favour), he cannot do so, because, the problem is sequential (cannot be parallelized, no matter how much hardware he provides) and will definitely take a fixed predictable amount of time to be solved. And by the time the hacker solves the problem and gets the result, the honest answer will also be calculated and the winner will have been already declared.

This is **PROOF OF DELAY**.

### Related work discussed
(jyothsna)
After identifying the environmental toll of traditional Proof of Work (PoW), our group spent today diving into the Related Work section of the Sprints paper. Before the authors introduce their own solution, they meticulously break down why existing "green" alternatives,specifically Proof of Stake (PoS),don't quite hit the mark if you want to maintain the gold standard of decentralization.

The Proof of Stake (PoS) Paradox : Most people point to Ethereum’s transition to Proof of Stake as the ultimate fix for blockchain's energy problem. By replacing physical hardware with locked digital tokens (stake), PoS reduces energy consumption by over 99%.However, the researchers argue that this efficiency comes with a significant hidden cost.

The Security Gap: PoW vs. PoS - The core of the paper’s argument lies in how a new user joins the network and determines which version of the blockchain is the true one. In Proof of Work (PoW), A node looks at the math. The chain with the most cumulative work is the truth.But, In Proof of Stake, A node must trust a set of checkpoints or existing peers to know the current state.PoW is Trustless (Math-based), but PoS is Trust-intensive (Social/Network-based).

Long-Range Attacks: Because creating a fake history in PoS costs almost nothing (it doesn't require electricity, just private keys from a long time ago), attackers can attempt to rewrite the entire history of the chain from the beginning.

Strict Network Assumptions: PoS requires much more stable, long-term network connectivity to remain secure. PoW is designed to work even in highly unstable or adversarial environments.

**The Sprints Mission: The Best of Both Worlds**
The takeaway from today’s reading is that the authors aren't satisfied with the trade-offs of PoS. They believe that the objective mathematical proof found in PoW is too valuable to throw away.Their goal is to create a protocol that keeps the mathematical truth of Bitcoin but achieves the carbon footprint of something much leaner. They want a system where a node can wake up, look at the math, and know the truth—all without burning down a forest to do it.

### Related work discussed
(*Hema*)
To reduce carbon emissions without abandoning the security of Proof of Work (PoW), Sprints re-engineers the mining timeline into a cyclic "heartbeat." Unlike Bitcoin’s continuous 24/7 marathon, Sprints uses a repeating cycle divided into two distinct stages: the Active Phase and the Sleeping Phase. This architectural shift allows mining hardware to remain powered down for the majority of the time, only "waking up" for brief, high-intensity intervals to secure the network.
During the Active Phase, miners run their ASIC hardware at full capacity to solve a standard PoW puzzle. This is the only window where energy-intensive computation occurs. Once a block is found, the network enters a mandated Sleeping Phase where rigs are switched off. To prevent desynchronization or attacks during this downtime, Sprints utilizes a Proof of Delay (PoD). The PoD is a cryptographic "timer" (a Verifiable Delay Function) that is easy to verify but must be solved sequentially. It acts as a mandatory speed limit, ensuring that a specific amount of real-world time passes before the next block can be mined.
By alternating these phases, Sprints maintains the rigorous security difficulty of Nakamoto Consensus while only being "on" for a fraction of the day. This disciplined scheduling allows for an order-of-magnitude reduction in the network’s operational carbon footprint, proving that sustainability can be achieved by structuring when mining occurs rather than replacing the mining process itself.
### Details of the solution proposed
*(shreekar)*
to actually fix the energy issue without ruining the security, the paper introduces their brand new protocol called "sprints". the core idea they came up with is something called intermittent mining. basically, instead of letting miners run their heavy hashing algorithms 100% of the time, sprints forces them to stop and solve a proof of delay (PoD) puzzle first. this PoD acts like a verifiable delay function (VDF) and takes a set amount of time to compute, ensuring a uniform delay across all miners. but the genius part is that because it is a purely sequential math problem, it cannot be sped up by running a million processors in parallel. running it uses almost zero electricity compared to standard mining. it basically forces the massive, power-hungry mining rigs to sit in a low power idle state while the delay function is ticking away. only after this delay is finished can they start the high-energy part.

*(Tavishi)*

Carrying on with how sprints work, once the delay period finishes, the miners can attach their transactions and start the heavy Proof of Work (PoW) hashing phase. This PoW phase is basically a very short "sprint" that only lasts for about 5% of the total block time. To make sure the network doesn't lag from verifying those slow PoD puzzles, the protocol uses a clever network engineering trick called "lazy validation". When a node receives a block, it verifies the fast PoW part first and forwards the block immediately to its peers to prevent network bottlenecks, and then verifies the slow PoD part in the background. The PoW acts as an
"anti-spam filter". The protocol also uses a dual difficulty adjustment algorithm. This uses the second moments of the interval probability distribution to ensure block times stay constant even as both VDF hardware and standard ASICs get faster over time. This keeps the 5% sprint to 95% delay ratio perfectly balanced.

### Gaps or loopholes in the solutions
(jyothsna)
While reading through the paper, our group noticed some physical loopholes and gaps that the intermittent mining design introduces. One big physical loophole is the severe damage to the mining rigs themselves. Standard ASICs(Application-Specific Integrated Circuits) are designed to be turned on and left running at a constant high temperature for years. As the sprints protocol forces machines to rapidly power up to maximum heat for the 5% sprint and then power down to idle for the 95% delay, it causes severe thermal cycling. This constant heating and cooling causes the internal silicon components, wire bonds, and solder joints to rapidly expand and contract. Over time, this thermal stress causes micro-cracks inside the chips. This type of damage can heavily reduce the lifespan of the ASIC miners, potentially dropping it from a normal 3 to 5 years down to maybe just 18 to 24 months, which hurts the miner's profitability.

*(Hema)*
The primary security concern with intermittent mining is that an attacker might ignore the "off" periods to gain a head start. Sprints prevents this through a Sequential Race enforced by Proof of Delay (PoD). By requiring a PoD to be solved before any Proof of Work (PoW) can begin, the protocol creates a cryptographic bottleneck. Since this delay is non-parallelizable, it acts as a mandatory "speed limit" that neither honest miners nor attackers can bypass.

Because every participant must wait through these identical delay periods, the time advantage an attacker might seek effectively "cancels out." This levels the playing field, ensuring that the longest chain is still determined by hashing power rather than the ability to skip downtime. The researchers proved that even if the network is "off" for 95% of the time, the security threshold only drops from 50% to 49%. This confirms that Sprints can achieve an order-of-magnitude reduction in carbon emissions while maintaining the robust, decentralized security of the original Nakamoto Consensus.
### Our thoughts on how these gaps can be bridged
*(shreekar)*
we brainstormed some ways the industry could bridge these gaps if the sprints protocol was actually deployed. to solve the thermal cycling hardware damage, we think mining farms would have to completely abandon normal air cooling fans and switch to dielectric immersion cooling. by submerging the ASIC miners completely in tanks of specialized non-conductive fluid, the temperature stays highly stable because the fluid holds the heat. even when the machines power up and down rapidly for the sprints, the fluid prevents the violent temperature swings, neutralizing thermal stress and extending the hardware lifespan significantly. for the rare earth metal supply chain issue, the industry needs to adopt strict circular economy standards using blockchain tracking itself. organizations are already developing supply chain tracking for rare earth materials. by putting the lifecycle of every ASIC miner on a ledger, manufacturers could mandate e-waste recycling of defunct hardware so we can recover the rare metals instead of constantly mining for new ones.
### Empirical evidence demonstrating the improvement of the solution
*(Tavishi)*
to wrap up our blog series, we want to highlight the actual empirical evidence the researchers provided to prove this isn't just theory. they actually tested this protocol to prove it works in real life by patching 100 bitcoin core nodes and running them on an emulated network with 100ms of latency (which perfectly mimics the real global bitcoin network). the math and empirical data showed that even with a really short 5% PoW window, the attacker threshold only dropped slightly from the theoretical 50% down to 49%, so it stays incredibly secure against 51% attacks.finally, they calculated the total carbon footprint using the CO2e metric. looking at the total lifecycle (which includes the carbon emitted from manufacturing all that extra hardware), the total systemic carbon footprint is reduced to just 9.2% of legacy bitcoin. this definitively proves that an order-of-magnitude environmental improvement is possible without losing nakamoto security.