Data and Validation Layer

Flex is a community-owned platform that rapidly boosts dApp user growth, ensuring the value generated directly benefits the Flex community.

The data and validation layer forms the cornerstone of the Flex Network, responsible for task execution, data processing, and primary validation.

Task assignment algorithm:

The task assignment process utilizes a complex algorithm that considers multiple factors to optimize network efficiency:

JavascriptcodeCopy codedef calculate_assignment_score(validator, task):
    C_t = task.complexity
    P_v = validator.performance_score
    A_v = validator.availability
    W_v = validator.stake_weight
    
    base_score = (P_v * A_v) / C_t
    reputation_factor = calculate_reputation_factor(validator)
    stake_factor = W_v ** 0.5
    
    return base_score * reputation_factor * stake_factor

def assign_task(task, validators):
    scores = [calculate_assignment_score(v, task) for v in validators]
    probabilities = softmax(scores)
    return np.random.choice(validators, p=probabilities)

The calculate_assignment_score function incorporates additional factors such as validator reputation and stake weight. The assign_task function uses a softmax distribution to introduce a degree of randomness while still favoring higher-scoring validators.

Peer validation determination:

The number of peer validators (N_t) required for a task is dynamically calculated based on the task's complexity and the network's current state:

$$N t ​ =max(⌈ C(t) ∑ v∈V ​ A(v) ​ ⌉,⌊log 2 ​ (∣V∣)+ C ˉ C(t) ​ ⌋)$$

where:

• V is the set of all validators

• A(v) is the availability of validator v

• Cˉ is the average task complexity

This formula ensures that more complex tasks or tasks during periods of low validator availability receive more peer validations.

We invite developers, researchers, and enthusiasts to work with our contributors for this section. Start contributing here.