Batch Operations

Batch operations allow you to perform multiple KiviGo operations efficiently in a single call. This is particularly useful for improving performance when dealing with large datasets or when you need to maintain consistency across multiple operations.

Overview

KiviGo provides three types of batch operations:

• BatchSet: Set multiple key-value pairs at once
• BatchGet: Retrieve multiple values by their keys
• BatchDelete: Delete multiple keys at once

Backend Support

Not all backends support batch operations. Check your backend's documentation:

Backend	Batch Support	Implementation
BadgerDB	✅	Native transactions
BoltDB	✅	Native transactions
Redis	✅	Pipeline operations
Consul	✅	Transaction API
etcd	✅	Transaction API
MongoDB	✅	Bulk operations
MySQL	✅	Prepared statements
PostgreSQL	✅	Prepared statements
Memcached	✅	Multi-operations

Basic Batch Operations

Batch Set

Store multiple key-value pairs at once:

package main

import (
    "context"
    "log"
    
    "github.com/azrod/kivigo"
    "github.com/azrod/kivigo/backend/redis"
)

func main() {
    // Setup client
    kvStore, err := redis.New(redis.DefaultOptions())
    if err != nil {
        log.Fatal(err)
    }
    defer kvStore.Close()
    
    client, err := kivigo.New(kvStore)
    if err != nil {
        log.Fatal(err)
    }
    
    ctx := context.Background()
    
    // Prepare batch data
    batchData := map[string]interface{}{
        "user:1": map[string]string{
            "name":  "Alice",
            "email": "alice@example.com",
            "role":  "admin",
        },
        "user:2": map[string]string{
            "name":  "Bob", 
            "email": "bob@example.com",
            "role":  "user",
        },
        "user:3": map[string]string{
            "name":  "Charlie",
            "email": "charlie@example.com", 
            "role":  "user",
        },
        "config:database": map[string]interface{}{
            "host":    "localhost",
            "port":    5432,
            "ssl":     true,
            "timeout": 30,
        },
    }
    
    // Perform batch set
    err = client.BatchSet(ctx, batchData)
    if err != nil {
        log.Fatal(err)
    }
    
    log.Println("Batch set completed successfully")
}

Batch Get

Retrieve multiple values by their keys:

func batchGetExample() {
    // ... setup client ...
    
    ctx := context.Background()
    
    // Keys to retrieve
    keys := []string{
        "user:1",
        "user:2", 
        "user:3",
        "config:database",
    }
    
    // Perform batch get
    results, err := client.BatchGet(ctx, keys)
    if err != nil {
        log.Fatal(err)
    }
    
    // Process results
    for key, value := range results {
        fmt.Printf("Key: %s, Value: %v\n", key, value)
    }
    
    // Note: Missing keys are not included in results
    // Check if a specific key was found
    if userData, found := results["user:1"]; found {
        var user map[string]string
        err := client.Get(ctx, "user:1", &user) // Individual get for type safety
        if err != nil {
            log.Printf("Error decoding user:1: %v", err)
        } else {
            fmt.Printf("User 1: %+v\n", user)
        }
    } else {
        fmt.Println("user:1 not found")
    }
}

Batch Delete

Delete multiple keys at once:

func batchDeleteExample() {
    // ... setup client ...
    
    ctx := context.Background()
    
    // Keys to delete
    keysToDelete := []string{
        "user:1",
        "user:2",
        "user:3",
        "temp:session:abc123",
        "temp:session:def456",
    }
    
    // Perform batch delete
    err := client.BatchDelete(ctx, keysToDelete)
    if err != nil {
        log.Fatal(err)
    }
    
    log.Printf("Deleted %d keys", len(keysToDelete))
}

Advanced Batch Patterns

Batch Update Pattern

Update multiple records efficiently:

type User struct {
    ID       int    `json:"id"`
    Name     string `json:"name"`
    Email    string `json:"email"`
    LastSeen string `json:"last_seen"`
}

func updateUserLastSeen(client client.Client, userIDs []int) error {
    ctx := context.Background()
    currentTime := time.Now().Format(time.RFC3339)
    
    // Step 1: Batch get existing users
    keys := make([]string, len(userIDs))
    for i, id := range userIDs {
        keys[i] = fmt.Sprintf("user:%d", id)
    }
    
    results, err := client.BatchGet(ctx, keys)
    if err != nil {
        return fmt.Errorf("failed to batch get users: %w", err)
    }
    
    // Step 2: Update last seen time
    updates := make(map[string]interface{})
    for key, _ := range results {
        // Get the user individually for proper type handling
        var user User
        err := client.Get(ctx, key, &user)
        if err != nil {
            log.Printf("Failed to decode user %s: %v", key, err)
            continue
        }
        
        // Update last seen
        user.LastSeen = currentTime
        updates[key] = user
    }
    
    // Step 3: Batch set updated users
    if len(updates) > 0 {
        err = client.BatchSet(ctx, updates)
        if err != nil {
            return fmt.Errorf("failed to batch update users: %w", err)
        }
    }
    
    return nil
}

Batch Migration Pattern

Migrate data between different key formats:

func migrateUserKeys(client client.Client) error {
    ctx := context.Background()
    
    // Step 1: List old format keys
    oldKeys, err := client.List(ctx, "user_")
    if err != nil {
        return err
    }
    
    if len(oldKeys) == 0 {
        return nil // No migration needed
    }
    
    // Step 2: Batch get old data
    oldData, err := client.BatchGet(ctx, oldKeys)
    if err != nil {
        return fmt.Errorf("failed to get old data: %w", err)
    }
    
    // Step 3: Prepare new format data
    newData := make(map[string]interface{})
    for oldKey, _ := range oldData {
        // Extract ID from old key format (user_123 -> 123)
        idStr := strings.TrimPrefix(oldKey, "user_")
        
        // Create new key format (user:123)
        newKey := fmt.Sprintf("user:%s", idStr)
        
        // Get the actual data for type-safe handling
        var userData interface{}
        err := client.Get(ctx, oldKey, &userData)
        if err != nil {
            log.Printf("Failed to decode %s: %v", oldKey, err)
            continue
        }
        
        newData[newKey] = userData
    }
    
    // Step 4: Batch set new format data
    if len(newData) > 0 {
        err = client.BatchSet(ctx, newData)
        if err != nil {
            return fmt.Errorf("failed to set new data: %w", err)
        }
    }
    
    // Step 5: Batch delete old format data
    err = client.BatchDelete(ctx, oldKeys)
    if err != nil {
        return fmt.Errorf("failed to delete old data: %w", err)
    }
    
    log.Printf("Migrated %d users from old to new key format", len(oldData))
    return nil
}

Batch Validation Pattern

Validate and process multiple items:

func processUserBatch(client client.Client, users []User) error {
    ctx := context.Background()
    
    // Step 1: Validate all users
    validUsers := make(map[string]interface{})
    var errors []error
    
    for _, user := range users {
        if err := validateUser(user); err != nil {
            errors = append(errors, fmt.Errorf("user %d: %w", user.ID, err))
            continue
        }
        
        key := fmt.Sprintf("user:%d", user.ID)
        validUsers[key] = user
    }
    
    // Return early if any validation failed
    if len(errors) > 0 {
        return fmt.Errorf("validation errors: %v", errors)
    }
    
    // Step 2: Check for existing users (prevent overwrites)
    keys := make([]string, 0, len(validUsers))
    for key := range validUsers {
        keys = append(keys, key)
    }
    
    existing, err := client.BatchGet(ctx, keys)
    if err != nil {
        return fmt.Errorf("failed to check existing users: %w", err)
    }
    
    if len(existing) > 0 {
        existingKeys := make([]string, 0, len(existing))
        for key := range existing {
            existingKeys = append(existingKeys, key)
        }
        return fmt.Errorf("users already exist: %v", existingKeys)
    }
    
    // Step 3: Batch create users
    err = client.BatchSet(ctx, validUsers)
    if err != nil {
        return fmt.Errorf("failed to create users: %w", err)
    }
    
    log.Printf("Successfully created %d users", len(validUsers))
    return nil
}

func validateUser(user User) error {
    if user.ID <= 0 {
        return errors.New("invalid ID")
    }
    if user.Name == "" {
        return errors.New("name is required")
    }
    if user.Email == "" {
        return errors.New("email is required")
    }
    return nil
}

Performance Considerations

Batch Size Optimization

Different backends have different optimal batch sizes:

const (
    RedisOptimalBatchSize      = 1000
    BadgerDBOptimalBatchSize   = 500
    ConsulOptimalBatchSize     = 100
    DatabaseOptimalBatchSize   = 100
)

func processBatchesOptimally(client client.Client, data map[string]interface{}, batchSize int) error {
    ctx := context.Background()
    
    // Split data into optimal batches
    keys := make([]string, 0, len(data))
    for key := range data {
        keys = append(keys, key)
    }
    
    for i := 0; i < len(keys); i += batchSize {
        end := i + batchSize
        if end > len(keys) {
            end = len(keys)
        }
        
        batch := make(map[string]interface{})
        for _, key := range keys[i:end] {
            batch[key] = data[key]
        }
        
        err := client.BatchSet(ctx, batch)
        if err != nil {
            return fmt.Errorf("batch %d-%d failed: %w", i, end-1, err)
        }
        
        log.Printf("Processed batch %d-%d", i, end-1)
    }
    
    return nil
}

Monitoring Batch Performance

Track batch operation performance:

func monitoredBatchSet(client client.Client, data map[string]interface{}) error {
    ctx := context.Background()
    
    start := time.Now()
    size := len(data)
    
    err := client.BatchSet(ctx, data)
    
    duration := time.Since(start)
    rate := float64(size) / duration.Seconds()
    
    log.Printf("Batch set: %d items in %v (%.2f items/sec)", 
               size, duration, rate)
    
    if err != nil {
        log.Printf("Batch set failed: %v", err)
        return err
    }
    
    // Alert on slow batches
    if duration > 5*time.Second {
        log.Printf("WARNING: Slow batch operation detected: %v", duration)
    }
    
    return nil
}

Error Handling in Batch Operations

Partial Failure Handling

Handle partial failures gracefully:

func resilientBatchSet(client client.Client, data map[string]interface{}) error {
    ctx := context.Background()
    
    // Try batch operation first
    err := client.BatchSet(ctx, data)
    if err == nil {
        return nil // Success
    }
    
    log.Printf("Batch operation failed, falling back to individual operations: %v", err)
    
    // Fall back to individual operations
    var individualErrors []error
    successCount := 0
    
    for key, value := range data {
        err := client.Set(ctx, key, value)
        if err != nil {
            individualErrors = append(individualErrors, 
                fmt.Errorf("key %s: %w", key, err))
        } else {
            successCount++
        }
    }
    
    log.Printf("Individual operations: %d success, %d failed", 
               successCount, len(individualErrors))
    
    if len(individualErrors) > 0 {
        return fmt.Errorf("partial failure: %v", individualErrors)
    }
    
    return nil
}

Retry Logic for Batches

Implement retry logic for failed batches:

func retryableBatchSet(client client.Client, data map[string]interface{}, maxRetries int) error {
    ctx := context.Background()
    
    for attempt := 0; attempt <= maxRetries; attempt++ {
        err := client.BatchSet(ctx, data)
        if err == nil {
            return nil
        }
        
        if attempt < maxRetries {
            backoff := time.Duration(attempt+1) * time.Second
            log.Printf("Batch attempt %d failed, retrying in %v: %v", 
                       attempt+1, backoff, err)
            time.Sleep(backoff)
        } else {
            return fmt.Errorf("batch failed after %d attempts: %w", maxRetries+1, err)
        }
    }
    
    return nil
}

Best Practices

1. Use Appropriate Batch Sizes: Test different batch sizes to find the optimal performance
2. Handle Partial Failures: Implement fallback strategies for partial failures
3. Monitor Performance: Track batch operation metrics
4. Validate Before Batch: Validate all data before performing batch operations
5. Use Transactions When Available: Some backends provide transaction guarantees for batches
6. Consider Memory Usage: Large batches consume more memory
7. Implement Timeouts: Use context timeouts for batch operations

Batch operations are powerful tools for improving performance and maintaining consistency when working with multiple key-value pairs in KiviGo.