}

    /**
     * Get link speed in Mbps
     *
     * @return link speed
     */
    public int getLinkSpeed() {
        return linkSpeed;
    }

    /**
     * Set link speed in Mbps
     *
     * @param linkSpeed link speed
     */
    public void setLinkSpeed(int linkSpeed) {
        this.linkSpeed = linkSpeed;
    }

time scanning up to a sixteenth of the namespace. Object versions marked `PENDING` or `FAILED` are re-queued for replication.

Replication speed depends on the cluster load, number of objects in the object store as well as storage speed. In addition, any bandwidth limits set via `mc admin bucket remote add` could also contribute to replication speed. The number of workers used for replication defaults to 100. Based on network bandwidth and system load, the number of workers used in replication...

                while (addresses.hasMoreElements()) {
                    InetAddress addr = addresses.nextElement();
                    
                    // Estimate link speed (would need platform-specific code for actual speed)
                    int linkSpeed = ni.isVirtual() ? 100 : 1000;  // Default 1Gbps
                    
                    NetworkInterfaceInfo info = new NetworkInterfaceInfo(addr, linkSpeed);

 *       <i>deeply</i> immutable.
 * </ul>
 *
 * <h4>Performance notes</h4>
 *
 * <ul>
 *   <li>Implementations can be generally assumed to prioritize memory efficiency, then speed of
 *       access, and lastly speed of creation.
 *   <li>The {@code copyOf} methods will sometimes recognize that the actual copy operation is
 *       unnecessary; for example, {@code copyOf(copyOf(anArrayList))} should copy the data only

 *       <i>deeply</i> immutable.
 * </ul>
 *
 * <h4>Performance notes</h4>
 *
 * <ul>
 *   <li>Implementations can be generally assumed to prioritize memory efficiency, then speed of
 *       access, and lastly speed of creation.
 *   <li>The {@code copyOf} methods will sometimes recognize that the actual copy operation is
 *       unnecessary; for example, {@code copyOf(copyOf(anArrayList))} should copy the data only

   *       the integral of our function, between X permits and X-K permits, assuming we want to
   *       spend K saved permits.
   * </ol>
   *
   * <p>In summary, the time it takes to move to the left (spend K permits), is equal to the area of
   * the function of width == K.
   *
   * <p>Assuming we have saturated demand, the time to go from maxPermits to thresholdPermits is

                while (addresses.hasMoreElements()) {
                    InetAddress addr = addresses.nextElement();

                    // Estimate link speed (would need platform-specific code for actual speed)
                    int linkSpeed = ni.isVirtual() ? 100 : 1000; // Default 1Gbps

                    NetworkInterfaceInfo info = new NetworkInterfaceInfo(addr, linkSpeed);

type StandardClaims struct {
	AccessKey string `json:"accessKey,omitempty"`
	jwtgo.StandardClaims
}

// UnmarshalJSON provides custom JSON unmarshal.
// This is mainly implemented for speed.
func (c *StandardClaims) UnmarshalJSON(b []byte) (err error) {
	return jsonparser.ObjectEach(b, func(key []byte, value []byte, dataType jsonparser.ValueType, _ int) error {
		if len(key) == 0 {
			return nil
		}

The key features are:

* **Fast**: Very high performance, on par with **NodeJS** and **Go** (thanks to Starlette and Pydantic). [One of the fastest Python frameworks available](#performance).
* **Fast to code**: Increase the speed to develop features by about 200% to 300%. *
* **Fewer bugs**: Reduce about 40% of human (developer) induced errors. *

- Modify directory listing operations for caching
- Update change notification handling

---

### Phase 5: RDMA (SMB Direct) Support
**Priority: LOW** | **Estimated Effort: 8-10 weeks**

RDMA provides high-speed, low-latency data transfer for supported network adapters.

#### 5.1 Core RDMA Infrastructure
```
Package: jcifs.internal.smb2.rdma
├── RdmaTransport.java           - RDMA transport implementation